The Discretionary Powers of the Supreme Court

To better understand the Court's political nature, legal scholars have long been interested in the different choices judges make. While studies of the SCC are now fairly far reaching, researchers interested in the Court as a political institution have largely focused on three components of its discretionary power: (1) the power to decide a case, (2) the power to decide whether to grant leave and (3) the power to decide whether to grant standing or intervenor status. A brief summary of how these discretionary powers are exercised and their political consequences follows.

First, the SCC's decision-making power is undoubtedly the most consequential of these discretionary powers. Unsurprisingly, then, it is the one that has received the most scholarly attention, particularly as concerns what factors may influence a judge's decision (Hausegger and Haynie, Reference Hausegger and Haynie2003; Hausegger et al., Reference Hausegger, Riddell and Hennigar2013; Johnson and Songer, Reference Johnson and Songer2009; Manfredi, Reference Manfredi, James, Abelson and Lusztig2002; Ostberg, Reference Ostberg and Wetstein2007; Radmilovic, Reference Radmilovic2010, Reference Radmilovic2013; Songer, Reference Songer2008; Songer et al., Reference Songer, Johnson, Ostberg and Wetstein2012; Tate and Sittiwong, Reference Tate and Sittiwong1989). While a range of different factors have been explored to better understand judicial decision making, personal ideology and policy preferences are the most frequently studied (Macfarlane, Reference Macfarlane2013: 15–38; Songer et al., Reference Songer, Johnson, Ostberg and Wetstein2012: 40–70). For the SCC, such research has established that while justices’ ideological preferences are reflected in their decision making, in comparison to the United States Supreme Court, for example, they are less likely to adhere to rote liberal and conservative stances. Instead, SCC justices’ voting behaviour appears to reflect a more nuanced, multidimensional approach (Ostberg, Reference Ostberg and Wetstein2007; Songer et al., Reference Songer, Johnson, Ostberg and Wetstein2012).

A second component of the Court's discretionary powers is the power to grant leave. For every case that is appealed to the SCC, one of the first choices the Court must make is whether to actually hear the case. Into the 1970s, this was hardly a choice at all, as the Court had little discretion over its docket. In fact, with the exception of civil suits involving disputes of less than $10,000, the SCC was required to hear all cases that were appealed to it. Amendments to the Supreme Court Act in 1975 gave the Court considerable control over its docket and, consequently, considerable discretion over the types of cases it now hears (Flemming, Reference Flemming2004: 7–9).Footnote ¹ Combined with the introduction of the Canadian Charter of Rights and Freedoms in 1982, the shift away from automatic appeals has had a profound impact on the work of the SCC, transforming it from a court primarily concerned with resolving private disputes to one of public law (Songer, Reference Songer2008).

Finally, another major discretionary power of the SCC comes from the ability to grant standing or choose who will participate in the cases it hears. For interest groups wanting to participate in a case, the most common route is to request intervenor status from the Court. If granted, this gives a third party the right to participate in the case proceedings and to provide comment on the legal issues being considered. Like the power to grant leave, the SCC's approach to granting intervenor status has transformed considerably over time. While would-be interest-group intervenors were often turned away in the early years of the Charter, by 1987 the Court began taking a more generous approach and now grants intervenor status to nearly all those who apply (Brodie, Reference Brodie2002: 36–40; Songer, Reference Songer2008: 124–6). While opening the doors to more intervenors is undoubtedly an important development as it relates to issues of representation, it also has significant consequences for the SCC's work: studies have found that who argues before the Court can affect the justices’ decision making (Manfredi, Reference Manfredi2004; Radmilovic, Reference Radmilovic2013). Relatedly, the relative ease by which interest groups are now able to participate in SCC cases is often presented as a critical development in the promotion of their political and policy interests (Epp, Reference Epp1998; Morton and Knopff, Reference Morton and Knopff2000).

Costs awarding is one of the SCC's discretionary powers that has received less attention from political scientists. It is now established practice in Canada that a party who succeeds in court can reasonably expect to have some of their legal expenses covered by the unsuccessful party through costs awards (Tollefson, Reference Tollefson2008–2009: 183–4). This model of costs awards has been labelled the two-way regime (Tollefson, Reference Tollefson2008–2009: 184). Costs are usually awarded on a “party and party” basis to partially indemnify the winning party (Orkin, Reference Orkin2010).Footnote ² Less common are costs awards on a “solicitor–client” basis, on a “full indemnity” basis, or orders of costs throughout, which allow for more substantial indemnification (Orkin, Reference Orkin2010). While the proportion of the costs defrayed varies by jurisdiction, the purpose of awarding costs is the same: to partially indemnify the successful litigant, deter frivolous actions and defences, encourage both parties to deliver reasonable offers to settle and discourage improper or unnecessary steps in litigation (Skidmore v. Blackmore, 1995: 339).

The SCC has near complete discretion over its costs orders. Under section 47 of the Supreme Court Act, the Court can “order the payment of the costs of the court appealed from, of the court of original jurisdiction, and of the appeal, or any part thereof, whether the judgment is affirmed, or is varied or reversed.” Typically, where an appeal to the SCC succeeds, costs will be awarded (Orkin, Reference Orkin2010). The Court's discretion to award costs in criminal cases, however, is more limited. The Criminal Code, 1985 (s. 683(3)) specifically prohibits costs awards in appeal cases dealing with indictable offences. Nonetheless, there have been some notable exceptions. Using a broad interpretation of section 24(1) of the Charter, some courts, including the SCC, have recognized the need to override these limitations to allow costs against the Crown in cases of Charter violations (Roach, Reference Roach1987: 265–7; Reference Roach2013: 514).

The SCC's discretion in awarding costs is important in terms of access to justice. At its core, access to justice aims to ensure that all those who require the assistance of the legal system can reasonably obtain it. In Canada, rising legal expenses have meant that access to justice has become an issue of increasing concern and one that is now frequently raised by the legal profession, including the Canadian Bar Association and the SCC (Cromwell, Reference Cromwell2012; Graham, Reference Graham2013; Makin, Reference Makin2011). As former Chief Justice Beverley McLachlin has noted: “Access to justice is quite simply critical. Unfortunately, many Canadian men and women find themselves unable, mainly for financial reasons, to access the Canadian justice system” (McLachlin, Reference McLachlin2007). Most recently, in Trial Lawyers Association of British Columbia v. British Columbia, 2014, the SCC cited access to justice concerns as justification for striking down British Columbia's system of court hearing fees as they created an “undue hardship” for litigants of limited financial means.

The problem of rising court expenses appears especially difficult for public interest litigation. As described by the SCC in British Columbia v. Okanagan Indian Band, 2003, public interest litigation includes those cases where there is a public benefit to be served through judicial resolution of the contested issues. While there are often compelling reasons for public interest litigants to pursue their cases, unlike private litigation, large damage awards are not an incentive to initiate actions. This has the potential to create a troubling financial barrier. As Berger (Reference Berger2002: 241) notes: “Important legal challenges that would vindicate rights and clarify fundamental freedoms are apt to be discouraged by the threat that a losing litigant will be burdened with the opposing party's costs.” The SCC's discretion over costs, then, has the potential to address, at least in part, the issue of access to justice for some of the most important public litigation cases in the country.

Understanding Costs Awards at the Supreme Court

The justices of the SCC have considerable discretion over how they exercise their judicial functions, including the awarding of costs. Their discretionary powers stem from the principle of judicial independence, which is constitutionally protected in Canada (Valente v. The Queen, 1985: 15). How these discretionary powers are used matters because they have the potential to affect who has access to and ultimately wins or loses at the SCC. On the particular issue of costs awards, the SCC may use its discretion to award (or not) costs to certain parties more than others. Practically speaking, this choice has consequences for litigants’ access to justice, but more broadly, it also provides a measure for how and to what extent the SCC may be prioritizing such access to justice concerns.

As already noted, the traditional approach to costs at the SCC has been that where an appeal to the SCC succeeds, costs are awarded to the successful party. However, the Court typically does not offer an explanation for its decision to award costs. This presents a challenge for understanding costs awards. One exception, however, is unconventional costs awards. In these cases, where the Court departs from its usual approach to costs, its decisions are very often accompanied by a stated rationale. An examination of these unconventional cases provides one of the only opportunities to analyze how the Court itself understands its exercise of awarding costs. To be clear, we are not suggesting that these exceptional cases on their own can provide a complete understanding of the SCC's approach to costs. Rather, by considering the Court's rationale for awarding costs in these cases, we should able to gain a fuller insight into the Court's approach to costs for public interest litigants more generally.

In Carter v. Canada, 2015, the SCC clarified the law surrounding the granting of “special costs.” Costs are usually awarded on a party-and-party basis, which only covers a portion of a litigant's legal expenses. In comparison, special costs cover all the legal expenses of a successful litigant. For public interest litigants, who face the prospect of what may be a long and inevitably costly legal battle, special costs represent a critical advance in access to justice. For the litigants in Carter, who succeeded in their legal challenge to have restrictions on physician-assisted death struck down as unconstitutional, the trial judge originally awarded special costs exceeding $1,000,000. By comparison, costs awarded on the usual party-and-party basis would not have exceeded $150,000 (Carter v. Canada, para. 134).

Beyond the Court's unanimous decision to award special costs to the successful litigants, the Carter ruling is significant for setting out a new test for determining when special costs should be ordered. As explained by the Court, costs should be paid on a full indemnity basis when two criteria are met: (1) public interest and (2) impecuniosity. Notably, the Court stressed that this test is intended to create a high standard so that costs orders departing from the usual rules will only be made in rare and exceptional cases. On this point, the majority noted that it is not enough for a case to transcend the individual interests of the successful litigant. The case must deal with a truly exceptional public interest matter that is of “significant and widespread societal impact” (Carter v. Canada, para. 140). Further, the plaintiff must not only demonstrate that they have “no personal, proprietary or pecuniary interest in the litigation that would justify the proceedings on economic grounds,” but also that they lack the financial capacity to bring the case forward by relying solely on private funding (Carter v. Canada, para. 140).

The SCC has also developed an approach for granting advance costs (also known as interim costs) to provide a party financial support while litigation is still ongoing and irrespective of the outcome of the case. The SCC first addressed advance costs in Okanagan Indian Band (2003), and the similarities between the test set out in this case and Carter are apparent. The majority in Okanagan Indian Band stated that advance costs can only be allocated in “rare and exceptional circumstances,” and subject to “stringent conditions and to the observance of appropriate procedural controls” (para. 1). The Court prescribed three necessary criteria for an application for an award of advance costs to succeed: (1) the party seeking interim costs cannot afford to pay for the litigation; (2) the claim to be adjudicated is prima facie meritorious; and (3) the issues raised transcend the individual interests of the individual litigant and are of public importance (British Colombia v. Okanagan Band, para. 40). With the exception of the criterion concerning whether the claim being adjudicated is meritorious—something that will have already been established in cases where full costs are awarded—the other two criteria, public importance and impecuniosity, are similar to those in Carter.

In Little Sisters Book and Art Emporium v. Canada (2007), a majority of the SCC added that “[i]n analysing the [Okanagan Indian Band] requirements, the court must decide, with a view to all the circumstances, whether the case is sufficiently special that it would be contrary to the interests of justice to deny the advance costs application” (para. 37). In the case at hand, the bookstore, which caters to the queer community, was denied advance costs when challenging Canadian Customs’ prohibition of four books ruled to be obscene. Despite special circumstances, the majority ruled that the case did not meet the merit and public importance requirements set out in Okanagan Indian Band. For the dissenting justices in Little Sisters Book and Art Emporium, the majority opinion was problematic: “[w]hether a case, though special, is not ‘special enough’ or fails to ‘rise to the level’ of compelling public importance is a subjective test whose outcome will inevitably depend to a significant extent on the eye of the beholder” (para. 154). The additional “special enough” standard appears to make advance costs more difficult to obtain than special costs.

Finally, in rare instances, the SCC has chosen to award costs to the nongovernment litigant even when they were unsuccessful (12 cases in the data considered here from 1970 to 2012).Footnote ³ While the Court provides little by way of explanation in this subset of cases, it is interesting to note that they largely appear to follow the criteria set out in Carter and Okanagan Indian Band of public interest and impecuniosity. For example, Borowski v. Canada (1989) dealt with the rights of the human fetus. At stake in Schachter v. Canada (1992) was discrimination between biological parents and adoptive parents regarding parental leave,Footnote ⁴ while A.C. v. Manitoba (2009) focused on whether a child has the right to refuse medical treatment for religious reasons. All these cases addressed legal issues with the potential to impact numerous individuals who were not parties in the legal challenges. The cases additionally imposed a considerable financial burden on the individuals mounting them. Notably, several of these cases were moot when costs were awarded.Footnote ⁵ In these instances, the original grievance had been solved by a legislative change, but the case continued through the legal system.

From a review of these select cases, the SCC appears to have developed a fairly clear approach to unconventional costs: the public interest litigant must be financially unable to support their legal actions alone and the case must address a legal issue of important public interest. In comparison, we know much less about the SCC's approach to awarding costs in public interest cases more generally. On this point, only a small body of research exists, with researchers generally focused on detailed case studies (Arvay and Latimer, Reference Arvay and Latimer2011; Fenrick, Reference Fenrick2007; Friedlander, Reference Friedlander1995; McCool, Reference McCool1996; Tollefson, Reference Tollefson1995, Reference Tollefson2002, Reference Tollefson2006, Reference Tollefson2011; Tollefson et al., Reference Tollefson, Gilliland and DeMarco2004). This is also true for studies looking at costs awards at other levels of court in Canada (for example, Emblem and Basmadjian, Reference Emblem and Basmadjian2016; Knutsen, Reference Knutsen2010; Mayeda, Reference Mayeda2011; Tollefson, Reference Tollefson2008–2009) or in other countries (for example, Campbell, Reference Campbell1983; Hodges et al., Reference Hodges, Vogenauer, Tulibacka and Cameron2010; Tollefson Reference Tollefson2008–2009, Reference Tollefson2011). These contributions provide compelling normative arguments outlining the importance of costs awards and key court cases in which the issue has been addressed. However, there is a near absence of research that analyzes general trends in costs awards.

To our knowledge, Lara Friedlander's (Reference Friedlander1994) study of noncriminal Charter cases from 1984 to 1993 (totaling 54 cases) is the only study to comprehensively analyze costs in constitutional cases. Here, Friedlander finds that the SCC strongly adheres to what she terms the “government rule.” With the “government rule,” when the government wins, the other party does not pay the government's costs as long as the government is substantially better able to absorb the costs of litigation and the other party has conducted the litigation responsibility and in good faith. If the government loses, it must pay the other party's costs (Friedlander, Reference Friedlander1994: 52). This approach to costs is consistent with the more commonly known “one way” model where public interest litigants do not pay costs if they lose but are compensated for their costs if they win (Tollefson, Reference Tollefson2009: 184). Of the 33 noncriminal Charter cases where one of the parties was a government body, Friedlander found that the Court followed the “government rule” 76 per cent of the time (Reference Friedlander1994: 52); however, it was followed less frequently where the government party was a specific arm of government (for example, British Columbia Securities Commission), as opposed to a representative of the government as a whole (for example, Attorney General of Canada) (Friedlander, Reference Friedlander1994: 53). Outside Friedlander's study, which is limited to noncriminal Charter cases and more than 20 years old, we have virtually no empirical data on the SCC's general approach to costs in public interest litigation. Consequently, there is a gap in the research for issues such as whether costs awards vary by type of legal case, by time period or by type of litigant. The data and analysis presented here aim to fill these gaps.

Data and Methods

Because we are interested in the role of costs in public interest litigation, SCC cases addressing issues of constitutional law were selected as the focus of our analysis. While it is certainly true that some nonconstitutional cases fall under the category of public interest litigation, cases of constitutional law are unique insofar as they often involve litigants who are attempting to determine their constitutional rights or other issues of broad significance. More than any other case type, constitutional cases are likely to present issues of public interest.

For this study, cases from 1970 to 2012 were analyzed. This 42-year period allowed trends in costs to be observed over time. To create this database of SCC constitutional law cases, the following sources were used. Data for this project were obtained primarily from the High Court Judicial Database (HCJD), which provides data on SCC decisions from 1969 to 2003 (Haynie et al., Reference Haynie, Sheehan, Songer and Tate2007). Additional cases from 2004 to 2012 were added and coded by the authors following the codebook of the HCJD. Constitutional cases were identified using the Lexum collection (Reference Lexum2015), which contains the SCC's judgments since 1875, by searching the subject “Constitutional law” on its website's advanced search option. These cases included those falling under the Canadian Bill of Rights, 1960, the Charter, the Constitutional Act, 1982 (excluding the Charter), “other constitutions,” which included the Constitution Act, 1867, the Indian Act, the Manitoba Act, 1870 and the North-West Territories Act, 1886, as well as federal and provincial human rights acts and codes, or a combination thereof. It should be noted that the Canadian Bill of Rights, 1960, the Indian Act and human rights legislations are only considered quasi-constitutional documents. While they implement constitutional imperatives by protecting rights and establishing institutional procedures, they can be amended without a formal constitutional amendment by a simple statutory vote (MacDonnell, Reference MacDonnell2016). Still, cases based on these statutes were included in the constitutional cases dataset because they were more likely to involve public interest litigation by those who face access to justice barriers. Finally, not all cases pertaining to the rights of Aboriginal peoples (section 35 of the Constitution Act, 1982) were captured using the “Constitutional law” category in the Lexum collection; thus, we used CanLII to identify these missing cases, which were then included in our dataset.

Each constitutional case assigned a case number by the SCC was considered separately, even if some had been indexed together in the reported Lexum collection decisions, as costs orders can vary by case. Furthermore, only those where one party is a government body and the party standing against it is a nongovernment body were selected. We included in the government category, the federal, provincial and municipal governments, as well as parapublic organizations, such as hospitals, school boards, educational facilities and law societies. In some cases, the government party was joined by an individual, organization or business. In other cases, the nongovernment party was represented by a Human Rights Commission. Each case was coded for whether the government and nongovernment parties were awarded costs. In total, 790 cases were coded.Footnote ⁶

Given the dearth of empirical research and the novelty of the data considered in this article, the analysis that follows is largely exploratory and descriptive. Accordingly, the first section of our analysis offers a snapshot of costs award trends over the study period and by case type. Nonetheless, we did have a number of expectations concerning what we would find. Following the research of Friedlander (Reference Friedlander1994), we expected that the SCC had continued to follow the “government rule” in noncriminal Charter cases. Specifically, we anticipated that nongovernment litigants were less likely to pay the costs of a government party when their case was unsuccessful, and that government parties were more likely to pay the costs of nongovernment litigants when their case was unsuccessful. We built on Friedlander's (Reference Friedlander1994) work by analyzing the extent to which the SCC adhered to the “government rule” during the period under study, both for Charter and other constitutional cases.

Using the notations in Table 1, Friedlander (Reference Friedlander1994) measured compliance with the “government rule” as follows:

$${\rm K =} \left( {{\rm A + D}} \right){\rm /}\left( {{\rm A + B + C + D}} \right)$$

While one of the objectives of this article is to determine whether the SCC was more likely to take public interest litigants’ needs into account than government parties in the awarding of costs, this dataset does include a wide variety of litigants. While some are undoubtedly persons or groups of limited resources who have taken on an exceptional financial burden to pursue litigation, a few are large companies that could hardly be considered prime examples for access to justice concerns. We recognize that the point where the cost to access the justice system becomes unreasonable is arguably subjective, which contributes to the infeasibility of measuring the financial capacity of each nongovernment litigant in the dataset. Nonetheless, we have identified different types of nongovernment litigants (described in findings), and consider whether this has an impact on how the SCC approaches the awarding of costs. We also distinguish between different types of government litigants (described in findings) and evaluate whether these different types affect how costs are awarded.

Table 1 Possible Outcomes of Public Litigation Cases

*Cases where the non-government litigant wins in part and loses in part are not considered here.

Findings and Analysis

Figure 1 shows the evolution of the order of costs from 1970 to 2012. The data show that the use of the SCC's discretionary power to award costs in constitutional cases has varied greatly over time from a high of 61 per cent of cases between 1981 and 1984 to a low of 16 per cent between 1993 and 1996. Altogether, the SCC ordered costs in 31 per cent of its constitutional cases. From 1970 to 1984, we observe an increase in the percentage of cases in which costs were awarded, but this percentage dropped dramatically from 1985 to 1996. Notably, although the Charter was implemented in 1982, the SCC's first Charter case, Law Society of Upper Canada v. Skapinker, was heard in 1984. The equality rights clause found in section 15 of the Charter,Footnote ⁷ on which many constitutional challenges are based, came into force three years after the Charter's introduction. Thus, this decrease in costs orders coincided with the beginning of the Court's Charter jurisprudence. During the pre-Charter period, costs were ordered in 53 per cent of cases, while during the post-Charter era, they were awarded in only 29 per cent of cases.

(all cases: N = 790; non-criminal cases: N = 384)

Figure 1 % Cases with Cost Awarded (1970–2012)

A dramatically different picture emerges, however, when criminal law cases are excluded from the analysis of costs (see Figure 1); the percentage of cases awarded costs increase dramatically to an average of 61 per cent. When criminal law cases were taken out of the equation, any post-Charter trends in costs disappear. Indeed, the difference between the two periods is only 1 percentage point (62% vs. 61%). Thus, the drop in post-Charter costs awards can largely be accounted for by the increase in the number of criminal law cases that accompanied the introduction of the Charter (Kelly, Reference Kelly2005; Songer, Reference Songer2008: 151–2).

Considering cases according to their issue type, constitutional cases dealing with criminal law made up the single largest number of cases. As expected, given the rules concerning costs in criminal law cases, costs were awarded in a very small number of cases, just below 4 per cent. More interesting are the cases dealing with civil rights and liberties, and public law, which together made up approximately 43 per cent of the cases examined here. Of the 148 cases in the category of civil rights and liberties, 55 per cent had costs awarded by the SCC. In contrast, of the 213 cases in the category of public law, 64 per cent had costs awarded by the court.

An examination of costs awards by type of constitutional challenge reveals that costs orders vary according to the constitutional basis of the case (noncriminal). Costs were ordered the least often in Bill of Rights, 1960 cases (11%), though this accounted for only nine cases. Since the SCC did not provide a costs order rationale in these cases, it is difficult to ascertain why these costs orders were low. Costs awarding is also lower for cases dealing with the Constitution Act, 1982 outside the Charter (27%). These 33 cases concerned Aboriginal rights litigation on section 35 grounds. In the 16 cases won by Aboriginal persons or groups, costs were awarded in six cases, whereas in the 13 cases won by the government actor, costs were awarded in two cases. Of the three cases in which both parties won in part, both parties were awarded costs in one case (Beckman v. Little Salmon/Carmacks First Nation, 2010).Footnote ⁸ Altogether, the number of cases for which costs were awarded was comparatively low but this finding presents an interesting contrast to the Okanagan Indian Band (2003) case, noted earlier, in which the Court took the unusual step of awarding advance costs. By contrast, in the 196 noncriminal law Charter cases, costs were ordered 59 per cent of the time. Similarly, noncriminal cases based on other constitutional documents and human rights acts and codes showed higher rates of costs orders, respectively 65 per cent of 153 cases and 69 per cent of 32 cases.

The breakdown of costs awards across case and constitutional type, however, only paints a partial picture. Of particular interest here is whether government or nongovernment parties are more likely to be the recipient of costs orders when they win or pay costs when they lose, which would suggest a particular pattern of behaviour by the SCC. To get a clearer picture of the SCC's approach to costs orders, it is necessary to consider the winners and losers in the cases surveyed. Recall that Friedlander found that the Court followed the “government rule” 76 per cent of the time in noncriminal Charter cases between 1984 and 1993 (Reference Friedlander1994: 52). Has this approach to costs continued?

The “government rule” was followed in 57.88 per cent of noncriminal constitutional cases between 1970 and 2012 (see Table 2). That is, nongovernment parties were less likely to pay costs when they lost and were more likely to receive costs when they won. Nevertheless, as Friedlander points out, it is possible that the SCC did not order unsuccessful nongovernment litigants to pay costs because the government party did not seek them (Friedlander, Reference Friedlander1994: 43–4; Tollefson, Reference Tollefson2008–2009: 186). Since the SCC does not provide this information in its judgment, it was not possible to control for this variable. Table 2 shows that the SCC's propensity to apply the “government rule” in noncriminal constitutional cases has varied over time. Between 1970 and 1984, the “government rule” was followed by the justices in 43.94 per cent of cases. In the post-Charter era, compliance with the rule increased to 60.61 per cent between 1985 and 1999 and to 61.86 per cent between 2000 and 2012.

Table 2 Compliance with the “Government Rule” in Non-Criminal Constitutional Cases Per Time Period (1970–2012)

^ap < .001

^bp < .01

^cp < .05

Note: This table does not include cases of reverse costs (N = 9), cases where both parties win in part (N = 23), or cases where the identity of the victor is not ascertained (N = 3).

In her research on noncriminal Charter cases, Friedlander found that between 1984 and 1993 the “government rule” was followed less frequently when the government party was a specific arm of government (67%), as opposed to a representative of the government as a whole (81%) (Friedlander Reference Friedlander1994: 53). When we look at noncriminal Charter cases for the same period, our findings do not follow Friedlander's. Compliance with the government rule was higher in the 27 cases where the government party was a specific arm of government (70.37%), compared to the 44 cases where it is a representative of the government as a whole (59.09%). However, neither of these findings achieve statistical significance. When examining the numbers from 1984 to 2012, no notable differences emerge. Compliance with the government rule was higher in the 53 cases where the government party was a specific arm of government (69.81%) (p < .05), compared to the 124 cases where it was a representative of the government as a whole (59.68%) (p < .05). These different results can be explained in part by differences in our datasets and, notably, the fact that Friedlander did not consider parapublic organizations as constituting governmental bodies. Thus, while we observed a small difference between government types, this finding is contrary to Friedlander's. Altogether, there is minimal evidence suggesting that the SCC demonstrates a different approach to costs awards depending on government type.

Another approach to observing differences in costs awarding behaviour is to consider the different types of nongovernment actors who participate in noncriminal constitutional cases. As Table 3 shows, we do observe differences between individuals, nongovernmental organizations and businesses, with the government rule being followed more frequently (66.13%) when the nongovernment actor is a nongovernmental organization. Assuming that on average nongovernmental organizations are in a weaker financial position than businesses to take on the costs of extended litigation, these findings suggest differences in approach on the part of the SCC.

Table 3 Compliance with the “Government Rule” in Non-Criminal Constitutional Cases According to Litigant Type (1970–2012)

^ap < .001

^bp < .01

^cp < .05

Note: This table does not include cases of reverse costs (N = 6), cases where both parties win in part (N = 23), cases where the nongovernment litigant is coded as “multiple players” (N = 27), or cases where the identity of the victor is not ascertained (N = 3).

Concluding Remarks

In this article, we explored an understudied component of the SCC's discretionary powers—costs awards. The findings presented here contribute to the larger body of scholarship focused on the political nature of the SCC. As has often been stated, the introduction of the Charter changed the nature of litigation in Canada. Costs awards appear to be another component of this change. While the government rule was followed in less than 50 per cent of noncriminal constitutional cases between 1970 and 1984, in the post-Charter era, compliance with the rule has increased to just over 60 per cent. Further, the rate by which the government rule is followed according to the different types of nongovernment litigants suggests a strategic approach by the Court that may be taking into account the significant costs that some individuals and groups take on in pursuing public interest litigation. In comparison to businesses, nongovernmental organizations were more likely to be awarded costs when they won and less likely to have to pay costs when they lost. By complying with the government rule in a majority of noncriminal constitutional cases, the judges have arguably lowered financial barriers and have facilitated the articulation of interests that may have difficulty finding an outlet in more traditional political channels. This view is further substantiated by the SCC's approach to unconventional costs awards.

However, the SCC's strategic approach to costs awards should not be overstated. While in comparison to government litigants, nongovernment litigants are more likely to have costs awarded when they win and less likely to have to pay costs if they lose, the numbers presented here are in the 60 per cent range. The SCC's approach to costs awards can hardly be viewed as a guarantee for public interest litigants considering the prospect of a long and expensive legal process. Although the SSC does appear to take litigants’ status into account when awarding costs, one can ask whether the Court goes far enough in addressing the costs of public interest litigants. The SCC's costs awarding power is important, but it can only be a weak reply to the powerful deterring effect of “fee reversal” on public interest litigation (Knutsen, Reference Knutsen2010).Footnote ⁹ The recent constitutional challenge by lawyer Aniz Alani to then Prime Minister Stephen Harper's refusal to fill Senate vacancies demonstrates the financial risks associated with public interest litigation (The Canadian Press, 2016). After the Liberal party of Justin Trudeau formed government in 2015 and resumed appointing senators, Alani's challenge was deemed moot and he was ordered by the Federal Court to pay costs to the federal government amounting to $20,489.52 (Federal Court of Canada 2017). In general, relying on the SCC to offset costs that have already been incurred is a maximum-risk strategy for public interest litigants since they still must absorb counsel and court fees up front, with no guarantees of relief, even if they win. Given the Court's inconsistent awarding of costs, fee reversal remains a strong deterrent to litigation.

This deterrent is one of the reasons that unconventional costs awards are so important as they relate to access to justice concerns. While the Court indicated in Carter that instances of special costs awards on a full indemnity basis for public interest litigants would only be made in exceptional circumstances, the Court's recognition of the special circumstances of public interest litigants is notable. Here, the Court's approach has been consistent with other scholarly findings on the SCC's approach to minority interests (Brodie, Reference Brodie2002; Manfredi Reference Manfredi2004). Moving forward, analyzing the Court's application of the special costs rules set out in Carter, and particularly how generous an approach it will take, will be an important issue for understanding the SCC's discretionary powers and its interests in public interest litigation.

While this article provides an extensive analysis of the SCC's approach to costs awards, much work remains. For example, archival work and interviews with both government and public interest litigants could help demystify to what degree and why governments decide not to seek costs in public interest cases. The importance of costs awards for public interest litigants and how they may promote or deter legal mobilization is also an important question for future study.