From Private Influence to Public Amendment? The Senate's Amendment Rate in the 41st, 42nd and 43rd Canadian Parliaments

Abstract

Recent reforms to the Canadian Senate removed senators from the Liberal Party caucus and changed the appointment process to be more nonpartisan. This article asks: to what extent did the reforms affect legislative oversight in the Senate? By studying the Senate's legislative amendments, I find that the reformed Senate is more willing to amend bills than it was previously. The reforms led to sharp increases in the Senate's amendment rate, the number of amendments moved and the percentage of successful motions in amendment. In interviews, senators revealed that they see oversight differently following the reforms. Senators no longer have opportunities to advise the government in caucus, so they have begun using amendments to exercise oversight. This article concludes that the reforms shifted senators’ understanding of their function of oversight, leading to a higher amendment rate and increased visible scrutiny of government by the Senate.

Résumé

Les récentes réformes du Sénat canadien ont retiré les sénateurs du caucus du Parti libéral et modifié le processus de nomination pour le rendre plus non partisan. Cet article pose la question suivante : dans quelle mesure les réformes ont-elles affecté le contrôle législatif au Sénat ? Il analyse les modifications législatives intervenues au Sénat et constate que le Sénat réformé est plus disposé à amender les projets de loi que les sénateurs ne l'étaient auparavant. Après les réformes, il y a eu de fortes augmentations du taux d'amendement, du nombre d'amendements proposés et du nombre de motions d'amendement réussies. Les entretiens avec les sénateurs révèlent qu'ils considèrent le contrôle différemment après les réformes. Les sénateurs ne perçoivent plus la possibilité de conseiller le gouvernement en caucus, et ils sont plus à l'aise pour utiliser les modifications afin d'exercer un contrôle. L'article conclut que les réformes ont modifié la compréhension qu'ont les sénateurs de leur fonction de contrôle. En outre, elles ont augmenté la visibilité du contrôle du Sénat après que le lieu du contrôle soit passé du caucus du gouvernement aux processus d'amendement formels.

Introduction

The Canadian Senate is a chamber of oversight, but the reasonable extent of that oversight is controversial. Critics argue that senators should never thwart government policy because they are unelected (Coyne, 2016). Yet Senate supporters contend that appointed senators can offer a tempered view of legislation without distraction from electoral pressures (Thomas, 2003). The Senate was active in its oversight of government following Confederation (Kunz, 1965; Mackay, 1963), but it presented fewer challenges to the government in the late twentieth and early twenty-first centuries (Heard, 2014; Macfarlane, 2019a). In 2014 and 2015, Justin Trudeau introduced Senate reforms to improve the chamber's oversight of government. He removed senators from the Liberal Party caucus and made the appointment process less partisan. The reforms were meant to increase the Senate's independence and reduce partisanship, allowing for better oversight of government.

To what extent have the reforms actually changed the Senate's function of oversight? To answer this question, I look at how often the Senate exercises oversight by amending legislation. I calculate the rate at which the Senate amended government bills in the 41st, 42nd and 43rd Parliaments (2011–2015, 2015–2019 and 2019–2021, respectively). I also count the number of amendments moved and adopted in the Senate during that time. The analysis finds that immediately following the reforms, the amendment rate was at a historic high and there was an unprecedented increase in amendments moved and adopted by the Senate. Independent senators moved more amendments to government bills after the reforms than government senators had moved before the reforms, suggesting that independent senators exercise stronger scrutiny over the government. Additionally, opposition amendments were more likely to be adopted following the reforms than they were before. These findings suggest that the Senate reforms effectively reduced partisanship and increased independence, but what does that mean for the quality of the Senate's oversight?

My interviews with senators clarify the nature of Senate oversight pre- and post-reform. The interviews revealed that the heightened amendment rate is the result of a shift in senators’ understanding of oversight following the reforms. The senators claimed that they had previously influenced bills in government caucus meetings, but those claims cannot be confirmed since those meetings are private. The interviews do make clear that senators’ perceptions of their job have changed since the reforms. They do not currently have access to the government caucus, so they no longer see the caucus as a potential venue of oversight. Now they consider amendments to be their main mechanism of oversight, so the Senate's oversight in the public arena has increased.

In brief, this article finds that the Senate reforms have changed the way senators perceive and exercise their function of oversight, leading them to amend more bills. The quantitative analysis shows an increase in Senate amendments following the reforms, suggesting that the reforms increased independence and reduced partisanship in the Senate. The qualitative analysis clarifies why the amendment rate increased: senators’ understanding of oversight has changed. Senators have no option to influence legislation in caucus (as they claim to have done in the past), and as a result, they now understand amendments to be a core part of their job. The turn toward amendment as a regular mechanism of oversight has improved public oversight by the Senate, allowing for better assessment of Canadian senators’ work in the public sphere.

Background Considerations

Sweeping reform of Canadian political institutions is unlikely due to the constraints of Canadian constitutional politics. Docherty (2002) predicted that any reform of the Canadian Senate would occur through small non-constitutional changes since the Canadian provinces have difficulty agreeing on broad constitutional reform packages. Indeed, the Senate reforms introduced by Trudeau in 2014 and 2015 worked outside the written text of the Constitution to improve the Senate's performance as a chamber of oversight. The reforms drew attention to the upper house in the mid-2010s, following a series of expense scandals that put the Senate under scrutiny.¹ On January 29, 2014, Liberal leader Trudeau removed all senators from the Liberal Party caucus (Cudmore, 2014). This gave senators more scope for independent action and removed them from the limited party discipline under which they had operated (Thomas, 2019). The senators ousted from the Liberal Party formed the Senate Liberal Caucus (SLC), which was not affiliated with the national party. After winning the election in 2015, Trudeau introduced the Independent Advisory Board for Senate Appointments (IABSA) in December of that year. The IABSA was designed to remove partisanship from the appointment process; now, a nonpartisan panel provides a shortlist of Senate candidates to the prime minister (Macfarlane, 2019b; Stos, 2017; Thomas, 2019). Out of 103 sitting senators at the end of the 42nd Parliament, 50 were appointed under the IABSA process. The reforms were meant to increase the Senate's independence and reduce its partisan nature to enhance oversight of government. This article assesses whether the reforms accomplished these goals.

The Senate has always been less partisan than the House of Commons because it is not the confidence chamber. It was designed to provide tempered oversight away from partisan activity, but partisanship in the Senate has increased in recent decades (Godbout, 2020). Russell (2001) observes that in parliaments dominated by partisan interests, second chambers provide a less partisan view on legislation. The recent reforms to the Canadian Senate attempted to align it with this principle of Westminster-style parliaments. To understand the reforms’ successes and failures, scholars are re-evaluating the Senate's function of oversight. As will be explored below, the literature on Senate oversight focuses on senators’ loyalty to the government in the House of Commons and party unity in the upper house.

Heller (2007) advances a view of bicameral legislatures as collaborative chambers that bargain over policies, eventually coming to an agreement. He calls this “bicameral bargaining” (248). This contrasts with the view of bicameral legislatures as competing chambers, where the house with more power gets its policy preferences realized (Heller, 2007). Bicameral competition has prevailed in Canada for years, and the executive in the House of Commons has usually gotten its way (Savoie, 1999). Unelected senators have been reticent to use their formal powers to amend, delay and veto bills from the elected house. As a result, Canada has operated under a weak form of bicameral competition, with a powerful House of Commons and disempowered Senate. But Heller's (2007) bicameral bargaining is possible within the rules of the Canadian Parliament if the Senate were to invoke its formal powers. If the Senate amends a bill, the House of Commons must reconsider the legislation and decide if it will accept, reject or amend the Senate's amendment. This begins a process of message-sending and bargaining between the houses, with both houses approving the final version of a bill. A shift from bicameral competition to bicameral bargaining would mark a change in the Senate's function of oversight. This article looks for evidence of bicameral bargaining in the form of Senate amendments to bills, as well as the House of Commons’ responses to those amendments.

Smith (2003) argues that bicameralism calls for accountability and review, noting that senators can amend, delay and veto bills. In doing so, they can frustrate the government's agenda and get the House of Commons to reconsider legislation. This article focuses on Senate amendments to government legislation as an exercise of oversight. In committee or at Third Reading, senators can amend bills to change policy implementation or expected outcomes. The amendment rate shows the Senate's willingness to question the government's policy agenda—in effect, to exercise oversight.

The Senate's amendment rate has varied, historically. Kunz (1965) and Mackay (1963) observe that the Senate was active in its oversight in the decades after Confederation. Between 1867 and 1960, the Senate amended 20 per cent of bills from the House of Commons (Mackay, 1963).² In the late twentieth and early twenty-first centuries, senators showed more deference to the House of Commons. Heard (2014) found that in the decades preceding the Senate reforms, the amendment rate was as low as 4 per cent (in the 34th Parliament, from 1988 to 1993) and as high as 11 per cent (in the 37th Parliament, from 2001 to 2004). The Senate has challenged the government in the past (for example, Franks, 2003; Smith, 2003), but overall it has been reluctant to use its formal powers to exercise regular oversight of government legislation.

Research has found that the Senate reforms have changed the upper house's operations and influence in some ways, but in other ways, continuity prevails (Bridgman, 2021; Godbout, 2020; Macfarlane, 2021; Smith, 2019; Thomas, 2019; VandenBeukel et al., 2021). Scholars of the Senate are asking whether the reforms accomplished their goal of improving the Senate's function of oversight. Macfarlane (2021), Bridgman (2021), Godbout (2020) and VandenBeukel et al. (2021) have conducted empirical studies on the reforms’ effects, revealing a renewed interest in the Canadian Senate following the reforms.

Macfarlane (2021) finds less cohesive party action in the Senate following the reforms. The government has to work much harder to pass legislation because they have no seats in the upper house. He observes that there is a “highly individualized and disaggregated environment” in the Senate post-reform (108). Senate policy making is now more fragmented because most senators do not belong to caucuses; briefings from public servants are usually given to small groups, or even to individual senators. Macfarlane's research highlights the complex, shifting environment of the new Senate; there is less co-ordination among senators (though there remains informal co-ordination, as well as co-ordination between bill sponsors and ministers’ offices). This raises questions about whether independent senators maintain unity with the government despite their fragmentation. Moreover, it raises questions about the extent of organized efforts to oversee and amend government legislation.

Similarly, Bridgman (2021) finds evidence of reduced partisanship among senators, pointing to increased lobbying activity following the reforms. He concludes that lobbyists disproportionately target independent senators because they are not bound by representative duties to political parties. There are few opportunities for outside actors to intervene in the political system, so lobbyists try to use independent senators to advance their policy agendas (since those senators have no prior commitment to any party's interests). Bridgman's (2021) findings cast doubt on the government's capacity to generate loyalty from independent senators and on its ability to pass its legislation through the Senate with little pushback. If senators are listening to actors other than the government, they may be more likely to exercise oversight based on lobbyists’ opinions.

By contrast, Godbout (2020) argues that claims of independence in the reformed Senate might be exaggerated. By analyzing legislative voting records of senators from the 1st to the 41st Parliaments, Godbout finds that the Senate had a high level of partisanship throughout the twentieth century. In his analysis of the 41st Parliament, Godbout observes that party voting unity in the Senate did not change much following the expulsion of senators from the Liberal caucus. VandenBeukel et al. (2021) found that party discipline declined in the Senate following the reforms, but independent senators were still most likely to vote with the government. Critics have wondered about the motivation for the Senate reforms, given that Trudeau removed senators from the caucus when he was the leader of the third-place party in the House of Commons and given that the independent senators he appointed have consistently voted with his government. When it comes to legislative voting in the Senate, Godbout (2020) observes that the reforms do not seem to have made sweeping changes to party unity. If independent senators rarely vote against the government, this calls into question any improvement in Senate oversight.

Godbout (2020) and VandenBeukel et al. (2021) focus on recorded votes, but does the pattern of independent senators’ loyalty to the government hold when votes are not recorded? This article looks at amendments in the Senate, which are normally passed by voice votes. Votes on motions in amendment may be recorded if a senator requests it, which usually happens when senators want to publicize their positions on controversial amendments. Votes can be held by a show of hands, but in those cases, only the numbers of yeas and nays are recorded, and not the senators’ parties. Motions can also be passed or defeated on division, but most of the time, the Journals of the Senate record votes on amendments as simply “adopted” or “negatived.” While those votes cannot be broken down by party, this article studies who proposed amendments and whose amendments are likely to be successful, taking partisan dynamics into account.

This article is informed by long-standing work on Senate oversight and by recent scholarship on the reformed Senate. In effect, contemporary scholars have asked whether the reforms improved the Senate's oversight of government. Macfarlane (2021), Bridgman (2021), Godbout (2020) and VandenBeukel et al. (2021) clarify the partisan dynamics of the reformed Senate, but questions remain about whether the reforms emboldened senators to amend more government legislation. Moreover, has Parliament shifted from bicameral competition to bicameral bargaining since the reforms? To establish whether the reforms have changed the Senate's function of oversight, I explore Senate amendments in the period around the reforms and I investigate senators’ perceptions of their oversight.

Amendments in the 41st, 42nd and 43rd Parliaments

Research design

This analysis concentrates on the Parliament immediately before the reforms (the 41st Parliament, 2011–2015) and the Parliaments following the reforms (the 42nd Parliament, 2015–2019, and the 43rd Parliament, 2019–2021). Government bills are the focus, as they address the government's policy agenda; amending those bills indicates senators’ disagreement with the government's plan. While private members’ bills and Senate public bills can address important policy issues, amending them does not usually present a challenge to the government's policy agenda.

To calculate the amendment rate in the 41st, 42nd and 43rd Parliaments, I used the LEGISinfo database, published online by the Library of Parliament. I counted whether bills were amended in committee or at Third Reading (or both).³ Bills that had amendments proposed in committee but were not passed with amendments by the Senate are not counted in the amendment rate. In this article, the amendment rate refers to the percentage of government bills passed by the Senate at Third Reading with amendments. For the amendment rate calculation, each amended bill is counted as one amended bill, regardless of the number of amendments to the bill.

I studied Hansard from both the House of Commons and the Senate to analyze messages between the houses about amendments. By doing this, I determined whether the House of Commons had accepted the Senate's amendments, rejected the amendments, or amended the amendments. In many cases, more than one of the above categories applies to each bill. I also noted the Senate's response to the House of Commons’ decisions. Tracking the House of Commons’ acceptance of Senate amendments allowed me to analyze whether the Senate exercised legislative oversight and to see whether the lower house engaged with the changes made by the upper house (indicating a shift toward bicameral bargaining).

I consulted Senate committee meeting minutes and the Journals of the Senate to determine which senators proposed amendments and which amendments were passed. There are several ways to count amendments, which makes the task complex. Feldman (2021) outlines strategies for counting amendments, including counting motions in amendment, per cent change of legislative text or even total length of amendments. He concludes that the method used to count amendments should be appropriate to the information sought. I have counted senators’ motions to amend bills, called “motions in amendment,” each of which represents an intent to make a policy change. Those motions in amendment contain subclauses, which include all the major and minor changes required to make a substantive change (for example, changing both the French and English versions of a bill, deleting a clause, and amending other clauses which refer to a deleted clause). Since the focus of this article is senators’ intent to challenge the government, I counted motions in amendment by senators in committee and at Third Reading, noting the movers’ names and parties. Each motion in amendment is counted as one attempt to amend by a senator. Each motion in amendment adopted by the Senate (or contained in a committee report adopted by the Senate) is counted as one amendment. If a senator moved the same motion in amendment both in committee and at Third Reading, they are counted as two separate motions in amendment (as they represent two attempts to exercise oversight and they are presented to different audiences). I used Microsoft Excel to organize and analyze the amendment data in a spreadsheet (see online appendix).

Results of amendment analysis

The Senate's amendment rate increased dramatically in the 42nd Parliament, following the reforms (see Figure 1). The data show that in the 41st Parliament, immediately before the reforms, the amendment rate was relatively low. Out of 122 government bills that received Royal Assent, the Senate amended eight of them, an amendment rate of 7 per cent. Following the reforms, the Senate amended 31 of 88 bills that received Royal Assent in the 42nd Parliament, an amendment rate of 35 per cent. The amendment rate in the 43rd Parliament was significantly lower than that of the 42nd Parliament. Of 31 bills that received Royal Assent, the Senate amended two of them, putting the amendment rate for the 43rd Parliament at 7 per cent. However, the 43rd Parliament was dominated by the COVID-19 pandemic; safety precautions severely restricted the number of committee meetings, which limited senators’ opportunities to move amendments. Due to the abnormal circumstances in the 43rd Parliament, this article focuses on the remarkably high amendment rate in the 42nd Parliament, which sat from 2015 to 2019, immediately following the reforms.

Figure 1. Government Bills That Received Royal Assent Amended by the Senate in the 41st, 42nd and 43rd Parliaments

During the 42nd Parliament, 28 government bills originated in the House of Commons and were amended by the Senate (the other three government bills amended by the Senate originated in the Senate). The House of Commons accepted amendments on 22 of the amended bills, rejected amendments on 18 of them, and amended the Senate's amendments on 14 of them. Notably, the Senate never insisted on its amendments, meaning it did not obstruct the will of the elected chamber. Rather, it presented the House of Commons with opportunities to re-examine its policy. In many cases, the government was open to the changes made.

The acceleration of the amendment rate is of note. Between the removal of senators from the Liberal caucus on January 29, 2014, and the end of the 41st Parliament on August 2, 2015, 56 bills received Royal Assent, one of which was amended. This is an amendment rate of 2 per cent. The senators who had previously sat in the Liberal caucus did not successfully amend more government legislation than they had in the past, once they were freed from party discipline. As more independent senators were appointed after Trudeau became prime minister, the amendment rate began to increase. Macfarlane (2019a) reported that in 2017, halfway through the 42nd Parliament, the Senate had amended 12 per cent of bills by that point. The data in this article show that by the end of the 42nd Parliament, the amendment rate had increased even further, to 35 per cent. The increase in the amendment rate coincides with the introduction of more independent senators as the 42nd Parliament continued.

Individual motions in amendment also increased substantially following the Senate reforms (see Figure 2). During the 41st Parliament, there were 111 motions in amendment, and the Senate passed 11 of them. During the 42nd Parliament, there were 640 motions in amendment and the Senate passed 495 of them. In the Parliament following the reforms, there was a 477 per cent increase in motions in amendment and a 4,400 per cent increase in Senate amendments.

Figure 2. Success of Amendments Moved in the 41st and 42nd Parliaments

In using the amendment rate as a proxy for senators’ willingness to check the government, I must account for the fact that the government sometimes amends its own legislation. In the 41st Parliament, senators in the Conservative Party of Canada's (CPC) government caucus made eight motions in amendment, all of which were adopted. Following the reforms, in the 42nd Parliament, the Government Representative Office (GRO) made 37 motions in amendment, and similarly, all of them were adopted. Moreover, several senators in the Independent Senators Group (ISG) and Senate Liberal Caucus (SLC) sponsored government bills. There is evidence in Hansard that the government asked some sponsors to amend their own legislation (and this reflects what I heard from senators in interviews). Usually, when that happens, the Senate respects the will of the government and agrees to include the amendment. Of the 495 amendments passed in the 42nd Parliament, 99 of them were moved by non-GRO senators who were sponsors of the bills (on a total of 13 bills). This leaves 396 amendments that likely came from outside the government in the 42nd Parliament. By comparison, there were three successful non-government amendments in the 41st Parliament. These data show that there was a sharp increase in amendments that did not come from the government.

The number of Official Opposition amendments has also increased following the reforms. In the 41st Parliament, Liberal senators moved 103 motions in amendment, and three of them were adopted. In the 42nd Parliament, Conservative senators moved 267 motions in amendment, and 165 of them were adopted. It is difficult to know how this particular cohort of Conservative senators would have behaved as opposition in an unreformed Senate, but opposition attempts at amendment have more than doubled since the Liberals were the opposition party. Moreover, the success rate of opposition amendments went from 3 per cent to 62 per cent, showing that the new cohort of independent senators is more willing to pass opposition amendments than government senators have been in the past.

The overall success rate of proposed amendments was 10 per cent in the 41st Parliament and 77 per cent in the 42nd Parliament. But when this is broken down by party, it is clear that some senators are more successful than others in passing amendments (see Figure 3). As mentioned, before the reforms, Conservative government senators’ motions in amendment were successful 100 per cent of the time (8 out of 8), and the opposition Liberal senators had a success rate of 3 per cent (3 out of 103). Following the reforms, the GRO's success rate for its motions in amendment was 100 per cent (37 out of 37), while ISG senators had a 90 per cent success rate (246 out of 272) and SLC senators had a 72 per cent success rate (38 out of 53). CPC senators had a slightly lower success rate for their motions in amendment, at 62 per cent (165 out of 267). Non-affiliated (NA) senators who were not in the GRO had a 50 per cent success rate (one out of two).

Figure 3. Success of Amendments Moved in the 42nd Parliament (by party/group)

Overall, there was a clear increase in the Senate's amendment rate following the reforms. Moreover, there was a strong increase in the number of amendments proposed and the percentage of them adopted. All parties and groups are proposing more amendments than they have in the past, with varying levels of success.

Discussion of amendment data

The high amendment rate following the reforms marks a dramatic increase from the pre-reform amendment rate. I interpret the sharp increase in Senate amendments to mean that senators publicly challenged the government more often following the Senate reforms. The analysis suggests that Senate independence increased and partisanship decreased. Independent senators introduced many more amendments during the 42nd Parliament than government senators did during the 41st Parliament, and opposition amendments were more likely to be passed. Moreover, the data show that the House of Commons considered the Senate's amendments and implemented many of them. This is consistent with a shift toward Heller's (2007) concept of bicameral bargaining. As discussed above, the emergence of bicameral bargaining following the Senate reforms suggests that the Senate began exercising more effective oversight in the public eye.

Conflict between the Houses of Parliament, such as Senate amendments to government bills, is more likely when there is a divided Parliament (wherein the House of Commons and the Senate are controlled by different parties) (Godbout, 2020; Kunz, 1965). Whether the 42nd Parliament can be considered a divided Parliament is debatable; the Liberal Party whip did not control any votes in the Senate, as it did in the House of Commons. So did independent senators act as challengers to the government? VandenBeukel et al. (2021) observe that independent senators demonstrated strong voting unity with the government in the 42nd Parliament, showing very little conflict and suggesting that Parliament was not divided. Here, I consider motions in amendment, most of which are passed by (unrecorded) voice vote. The analysis found that ISG senators proposed and passed a record number of amendments to government legislation. The data in this article suggest that senators did challenge the government through amendments, and there was more conflict between the Houses than previously thought. In other words, data from the Senate's unrecorded votes make it clear that the independent senators often participated in oversight of the government.

But is it that senators are exercising oversight? Or could they be fulfilling the government's wishes to amend its own legislation? Ninety-nine amendments came from sponsors of government bills, who would consult with the government during the process, and a further 37 amendments came from the GRO. Sponsors may present amendments to the government for approval, or the government may realize a problem after a bill has been passed in the House of Commons and ask the Senate to amend it. However, 359 successful amendments were not moved by bill sponsors, suggesting that there was more of an effort to challenge government policy following the Senate reforms. Moreover, the increased success rate of the opposition's motions in amendment (from 3 per cent to 62 per cent) suggests that ISG senators considered all challenges to the government and sided with the Conservatives on many of their amendments.

The high amendment rate may also be related to outside efforts to influence government legislation. Bridgman's (2021) work found that lobbyists disproportionately target independent senators, who have grown in number throughout the 42nd Parliament. Bridgman found an increase in lobbying of senators over the same period as I found a sharp increase in Senate amendments. It is difficult to ascertain how amendments originated, but future research could investigate whether there is any correlation between the number of lobbyists a senator meets and the number of amendments a senator proposes. Bridgman's findings may, in part, explain the high amendment rate in the 42nd Parliament observed in this article, as the independent senators played a prominent role in the introduction and passage of the high number of amendments.

The increase in amendments suggests that the Senate reforms have increased independence and decreased partisanship in the upper house. It is clear that senators have challenged the government more publicly now than in the past, but do amendments alone represent senators’ efforts to review legislation? In interviews, I spoke to senators about how they view their jobs as overseers and how they performed oversight before and after the reforms.

Interviews with Senators

Research design

As part of a larger project about senators’ representational behaviour, I interviewed sitting and former Canadian senators in the fall of 2019. I contacted the interviewees by email. I recruited interviewees based on a representative sample that was purposively selected according to senators’ sex, partisanship and regional representation. If a senator did not reply or was unable to speak with me, I contacted another senator who fit the same profile. I contacted 47 senators and two former senators. Five senators declined to be interviewed, and 21 senators did not respond to the request. I conducted a total of 23 interviews, with 21 sitting senators and two former senators. I spoke to a fairly representative sample of senators in terms of partisanship, though Liberal and non-affiliated senators were slightly overrepresented in my sample and Conservative senators were slightly underrepresented (see Table 1). I spoke to members of the leadership or facilitating team of each group in the Senate, as well as members of the GRO. Interviews were conducted mainly over the phone (three were conducted in person) and they lasted roughly 30 to 80 minutes. Senators spoke to me on the condition of confidentiality, and all but one agreed for the interview to be recorded on a digital recorder. The same senators who agreed to be recorded also agreed to be quoted without attribution. The recordings were transcribed at a later date.

Table 1. Partisan Characteristics of All Interviewees and Members of the Senate as of the End of the 42nd Parliament

Of the 23 senators with whom I spoke, 13 were appointed after the introduction of the IABSA in December 2015, so they sat as independent senators. Ten interviewees were appointed before the introduction of the IABSA, and nine of them had experience sitting in government caucuses: five senators sat in the Conservative caucus under Prime Minister Stephen Harper, and four senators sat in the Liberal caucus under Prime Ministers Paul Martin and Jean Chrétien (before senators were removed from the caucus by Justin Trudeau). One interviewee who served in the Conservative caucus and one who served in the Liberal caucus sat as non-affiliated senators in the 42nd Parliament, and Table 1 counts senators using their partisan status as of the end of that Parliament.

Analysis of interview data

The interview findings explain the increased amendment rate in the 42nd Parliament. As will be explored below, the reforms changed senators’ understanding of oversight. In the interviews, three senators claimed that they had more freedom to critique government bills after the reforms. Those senators see the high amendment rate as evidence that the reforms improved the Senate's legislative oversight. However, some senators appointed before the implementation of the IABSA offer a different interpretation: they claim that senators’ influence has not increased with the reforms so much as it has shifted venue. Senators with caucus experience claim that they used to influence policy in private through government caucus meetings. They contend that all senators (including their IABSA-appointed colleagues) used formal amendments to exercise oversight after they were excluded from the government caucus.

Five senators who had experience in government caucus claimed that before the reforms, they would critique bills privately, in caucus meetings (three of the other senators with caucus experience did not confirm or deny these claims, while one senator with experience in the Conservative caucus felt that dissent was not welcome within caucus). As a Conservative senator explained, “when we are in the government, we are also members of the national caucus. So, we will try to interact in the national caucus, and if [there is] something I think is wrong, we will work on the national caucus to try to amend the bill immediately in the House. We will start to make pushback there. So, we will not necessarily wait to have the bill in the Senate. So, we will work proactively to try to change the bill.”

Former Liberal and Conservative senators maintain that they would encourage ministers to change a bill in caucus before it was officially sent to the Senate. When Trudeau removed the senators from the Liberal Party caucus and appointed new senators as independents, they lost their perceived ability to suggest changes to legislation in Liberal caucus meetings. One former Liberal senator felt strongly that it was a mistake to remove senators from the government caucus: “We've lost, and the government has lost, and the prime minister has lost the opportunity to hear from seasoned, experienced parliamentarians in the Senate, who have less to gain by staying silent. The truth to power is more likely to come from a senator in the caucus than it would be from other members.” A Conservative senator observed that “the government cannot have the same access and the same point of view of the senators before the adoption of the legislation. And I think that's one of the reasons why you have seen more amendments with the last session [of Parliament], because the government didn't have access to this intelligence.” These interviewees argue that because amendments are a more public way to influence policy, the influence of senators has simply become more visible. Senators who were appointed before the introduction of the IABSA felt that the increased amendment rate was a result of senators’ oversight shifting from the government caucus to the amendment process in the public arena.

This point has been observed elsewhere by Conservative senators (Macfarlane, 2019b; The Agenda with Steve Paikin, 2019). In the interview research presented here, three senators who had sat in the Liberal caucus agreed with the argument that the amendment rate had likely increased because senators were making formal amendments to legislation rather than offering suggestions about legislation within caucus meetings. Senators with experience in both Conservative and Liberal governments agree that before the reforms, their main channel of oversight was through government caucus meetings.

Because caucus meetings are private, there is no way of verifying that the senators did have the influence in caucus that they claim to have held. Moreover, the mood and purpose of caucus meetings can vary widely depending on the prime minister. Marland (2020) details different caucusing styles among prime ministers, noting that some prime ministers are much more willing to accept suggestions from the caucus floor than others. This complicates senators’ claims about the effectiveness of their oversight in caucus. However, it is important that senators felt like they lost their main channel of influence when they were removed from the government caucus. With that channel of influence unequivocally closed, senators have limited opportunities to influence legislation, which might explain, in part, their turn towards amendments in the 42nd Parliament.

Another piece of the explanation for the increased amendment rate is that newly appointed senators see amendments as integral to their function of oversight. All of the senators I interviewed shared the view that their job is mainly oversight of government legislation. Senators appointed by Trudeau (since the introduction of the IABSA) indicated that they see amendments as a key part of their job. In the interview sample, every senator appointed by Trudeau viewed amendments as a way to pursue their policy agendas—in effect, to stand up for what they believe is right. One senator noted: “With respect to independent senators, I think our main activity is trying to improve bills through amendment.” This attitude among senators is reflective of their desire to improve legislation, though it is worth noting that any purported improvement is subject to senators’ own worldviews and policy priorities. Another ISG senator sees amendments as a form of advising the government, not overruling it: “We give [the government] advice, they take the decision.” This view is comparable to pre-IABSA senators’ sense of themselves as advisors. However, the interviews show that before the reforms, senators understood the advisory process to take place in private, during caucus meetings, rather than through amendments to legislation.

While newer senators see amending bills as a key component of their jobs, they are also respectful of the fact that the democratic will rests with the House of Commons. This is consistent with Heller's (2007) idea of bicameral bargaining, in that the Senate is not competing with the House of Commons; rather, senators understand themselves to be giving the government another perspective on legislation. But the dramatically increased amendment rate has raised alarm in some circles, with some pundits considering it a constitutional crisis. Andrew Coyne (2017) and Peter O'Neil (2017) claim, respectively, that by amending government legislation at such a high rate, senators attempted to overrule the will of the House. Coyne (2019) even expressed concern that the high number of amendments by the Senate brought Canada to the brink of a constitutional crisis whereby the unelected Senate could overrule the will of the elected chamber. He posited that the mere threat of a defeat in the Senate could stop the government from introducing a piece of legislation. Coyne and O'Neil understand Canada's Parliament to be in bicameral competition, whereby the Senate presents a threat to the House of Commons, rather than bicameral bargaining, where the two houses can disagree, bargain and then come to an agreement.

Despite the concerns of some political pundits, the new senators do not assert any right to overrule the government (consistent with bicameral bargaining). An ISG senator said: “We should do our job, which is study, try to amend it, and push for debates or for revisions and reviews, and that's it. . . . We have to go with what the democratically elected government decides.” All of the senators I interviewed agreed that it is not their place to stand in the way of legislation put forward by the democratically elected government in the normal course of legislative oversight. A senator in the ISG's facilitating team explained that independent senators sometimes felt that they had to vote for bills with which they did not agree; they knew they could count on Conservative votes against a Liberal government bill, and they did not want to contribute to “an accidental defeat of a government bill.” This might explain the discrepancy between VandenBeukel et al.'s (2021) finding that independent senators have strong voting unity with the government and this article's finding that the high number of amendments indicates tension between independent senators and the government. Newly appointed senators see their job as challenging the government by amending bills, but generally they defer to the elected government's will when it comes time to vote at Third Reading.

The interviews in this study reveal that the reforms prompted a shift in how senators conceptualize their roles as overseers. Consistent with Macfarlane's (2021) finding of a fragmented and disaggregated partisan environment in the reformed Senate, senators claim to have lost the streamlined oversight process within the government caucus. Senators appointed before the introduction of the IABSA claim that before the reforms, advising the government in private caucus meetings was their most effective way of performing their function of oversight. Now that senators no longer perceive the government caucus as a venue of influence, their best option to exercise oversight is to amend legislation. Formal amendment is a much more visible, if disaggregated, process. Senators’ claims of losing caucus influence are consistent with the finding noted above, that Conservative government senators moved just eight amendments during the 41st Parliament, while GRO senators and government bill sponsors moved 136 amendments in the 42nd Parliament. The data suggest that senators who are acting in concert with the government have turned to formal amendments now that they cannot exercise oversight in caucus. Like Macfarlane (2021), this article finds that the Senate reforms have produced a less partisan, more independent environment. In this environment, senators publicly criticize the government by amending their bills. Influencing legislation through caucus might have been quicker (if and when it did occur), but using the amendment process is a more transparent, public oversight procedure, meaning that academics, the media and citizens can undertake a more thorough study of senators’ activity on legislation.

Conclusion

The amendment data presented here indicate that the reforms increased independence, decreased partisanship and caused a shift toward more public oversight. To assess the effectiveness of the Senate reforms, recent literature focuses on party unity in the upper house and loyalty to the government. Scholars have analyzed whether the reforms have reduced partisanship, increased independence and, in turn, improved Senate oversight. In this article, the quantitative amendment analysis indicates that the reforms successfully reduced senators’ partisanship and loyalty to the government; senators are now more likely to amend government legislation and to pass amendments from across the aisle. Moreover, government senators in the 41st Parliament amended far fewer government bills than independent senators in the 42nd Parliament did. This suggests that the reforms effectively increased the Senate's independence from government.

In terms of whether the reforms improved Senate oversight, my interview analysis reveals that the quality of oversight shifted. Senators’ claims about pre-reform oversight behind closed doors complicate any general claim that the reforms have improved Senate oversight. Through interviews, I observed that senators have no opportunity to offer advice in the government caucus, so they have turned to amendments to exercise oversight. Thus, this article concludes that public oversight in the Senate has increased. In that respect, the reforms have created a more transparent legislative process, where senators’ claims about their legislative efforts can be confirmed more easily. The increase in amendments to government legislation means that senators’ work is more visible than it would be if it occurred behind the closed doors of a caucus meeting.

There has only been a Liberal government since the reforms, so it is uncertain whether the high Senate amendment rate will continue when a Conservative government takes power and senators regain access to the government caucus (or when Conservatives regain a majority in the Senate). Further research is warranted here, as the partisan dynamics in the Senate continue to shift following the reforms. More attention to senators’ amendment activity could have a considerable impact on public perceptions of the Senate's role in Canadian democracy.