University of Calgary
UofC Navigation

Ablawg

Subscribe to Ablawg feed
Updated: 13 min 27 sec ago

Learnings from the Demographic Data on Litigants Without Counsel

11 hours 26 min ago

By: John-Paul Boyd

PDF Version: Learnings from the Demographic Data on Litigants Without Counsel

The demographic information on litigants without counsel available to date reveals a number of interesting patterns: most litigants appear to be 40 years old and older, and people in that age range are involved in litigation at rates far higher than those in younger age groups; although most litigants have lower incomes, a significant number have incomes around or exceeding the average income; and, litigants’ often high incomes match their educational achievements, which often exceed the average. All of this information strikes me as potentially useful when designing services and reforming processes for litigants without counsel.

In her 2013 report Identifying and Meeting the Needs of Self-represented Litigants, Julie Macfarlane obtained demographic data from 230 litigants without counsel in Alberta, British Columbia and Ontario. The Nova Scotia Department of Justice surveyed 58 litigants without counsel for its 2004 report Self-represented Litigants in Nova Scotia. For their 2005 report BC Supreme Court Self-help Information Centre Initial Evaluation Report, John Malcolmson and Gayla Reid surveyed about three hundred of the centre’s users.

Age 

In Alberta, British Columbia and Ontario, litigants without counsel belong disproportionately to the 40 plus age group. In Nova Scotia, the 35 to 44 year old age group is disproportionately represented among litigants without counsel:

  • In the Macfarlane report, 45% of the litigants surveyed were 50 and older. 32% were age 40 to 50 and 14% were age 30 to 40.
  • In the Malcolmson and Reid report, 27% were 50 and older. 41% were age 40 to 49 and 25% were age 30 to 39.
  • In the Nova Scotia report, 13% were age 55 and older. 19% were age 45 to 54 and 45% were age 35 to 44.

Here’s a chart comparing the percentage of litigants without counsel in each age group with Statistics Canada census data from 2006 and 2011 on the percentage of the general population in each age group.

(I could only find the 2006 data in mid-decade age cohorts, and it’s a bit difficult to compare the findings in the Malcolmson and Reid report with the census data as a result).

Income

Although the lion’s share of litigants without counsel have incomes that are lower than average, the studies to date show that a significant number have incomes that are toward the middle and high ranges:

  • In the Macfarlane report, 40% of the litigants surveyed earned less than $30,000 per year. 17% earned $30,000 to $50,000, and a whopping 31% earned $50,000 or more.
  • In the Malcolmson and Reid report, 61% earned $24,000 or less per year. 22% earned between $24,000 and $36,000, and 16.7% earned in $36,000 or more.
  • In the Nova Scotia report, 60% earned $29,999 or less per year. 28% earned $30,000 to $59,999 and 10% were had incomes in excess of $60,000.
  • In An Evaluation of Alberta’s Parenting After Separation Seminars, a 1999 report by the Canadian Research Institute for Law and the Family, 1,180 participants were surveyed. 40% earned less than $33,754 per year in 2014 dollars. 31% earned between $33,755 and $60,757, and 29% earned more than $60,758 per year.

Here’s a chart, from Statistics Canada’s Table 202-0407, Income of Individuals by Sex, Age Group and Income Source, showing the average income by age group in 2011 dollars for: Alberta, British Columbia and Ontario combined; Alberta; British Columbia; and Nova Scotia.

(Note that the average incomes will likely vary between urban, rural and remote communities.)

Educational attainment

The reports also suggest that litigants without counsel tend to have taken more schooling than most Canadians:

  • In the Macfarlane report, 73% of the litigants surveyed earned had attended college or university. 16% had completed or partially completed high school.
  • In the Malcolmson and Reid report, 65% had attended college or university. 35% had completed or partially completed high school.
  • In the Nova Scotia report, 71% had attended college or university. 29% had completed or partially completed high school.
  • In the Canadian Research Institute study, 47% had attended college or university.  52% had completed or partially completed high school.
  • A survey of a small number of litigants for the 2009 Calgary Judicial District report of the Alberta Legal Services Mapping Project found that 50% had attended college or university and that 30% had completed or partially completed high school.

According to Statistics Canada’s 2011 Education in Canada: Attainment, Field of Study and Location of Study report, 64% of Canadians have a post-secondary qualification of some nature, 23% have graduated high school and 13% have neither. Here’s a chart from the 2011 census showing the educational attainments of people age 25 and over for: Alberta; British Columbia; Ontario; and Nova Scotia.

(Note that this data shows only completed diplomas, certificates and degrees, and thus excludes people who took a year or two of post-secondary education before leaving school. Note also that the average level of educational attainment will likely vary between urban, rural and remote communities.)

A few preliminary conclusions and a few tentative suggestions

The reports’ findings on age are interesting. First, according to the Statistics Canada data, Canadians’ peak earning years are between 45 and 55. This is probably helpful for people involved in a court proceeding, as the bubble of litigants noted by the Macfarlane and Malcolmson and Reid reports falls right in the midst of this age group. However, given that the reports also show that the significant majority of litigants without counsel are unrepresented because they can’t afford to hire a lawyer, it seems that these particular litigants cannot afford to retain counsel even at their peak earning years.

Second, the data on age suggests that some adjustments may be required on the part of public legal education groups, who may want to retool their publications to accommodate the declining visual acuity of their prime markets in their web and print offering. Further, as the baby boomers work their way through Statistics Canada’s age groups, other issues are foreseeable such as an increased need for public health services relating to competency, the redesign of courthouses to accommodate increasing numbers of people with lower levels of mobility, and an increase in the number of elderly Canadians living in poverty as divorces split fixed incomes. The nice people at CARP will tell you that the impact of divorce on seniors can be disastrous, particularly for women.

The reports’ findings on income also suggest the need to increase the income threshold for the provision of legal aid. To recap, 40% of the litigants in the Macfarlane report earned less than $30,000 per year, 61% in the Malcolmson and Reid report earned less than $24,000 per year, and 60% in the Nova Scotia report earned less than $29,999 per year. However, according to Canadian Lawyer magazine’s 2014 survey of legal fees:

  • the average contested divorce costs $10,406, or as much as $39,522;
  • the average two-day trial costs $12,769, or as much as $27,591; and,
  • the average five-day trial costs $26,591, or as much as $59,700.

Clearly lawyers are unaffordable for the majority of litigants without counsel. Nevertheless, the data also suggests that litigants can afford at least some services, and that litigants with mid to high incomes can afford even more. It seems to me that this data argues powerfully for both increasing eligibility for legal aid, so that fewer litigants are left without counsel, and lawyers to consider the unbundling of legal services, providing services at a flat rate or on a barter basis, and exploring other creative approaches to paid legal services than the billable hour. I’ve written about the adverse effect of the billable hour model of legal services elsewhere.

Further, given the vicious spiral that self-representation can lead to, which I’ve also written about previously, one wonders whether increasing the legal aid limits to provide more people with lawyers would in fact increase the rate of settlement, reduce the number of trials and result in a net savings to the justice system as whole. Just a thought.

Finally, the data on educational attainment makes me wonder what has happened to all of the people with high school diplomas or less. The national Action Committee on Access to Justice in Civil and Family Matters notes in its final report that only about 6.5% of legal problems ever make it to court, but it is unlikely in the extreme that so many of the people with high school diplomas or less are bundled into the 93.5% who manage to resolve their legal issues outside of court, especially when we know that for people with low incomes, legal issues tend not come one at a time but cluster and multiply into other areas of the law. Are these people simply never entering the formal justice system and abandoning their rights and entitlements? It seems to me that the providers of public legal assistance services need to examine their client population and ask how to better engage people with lower levels of educational attainment.

This post originally appeared on Access to Justice in Canada.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

Federal Court of Appeal Reviews CEAA “Justification” Determination for Lower Churchill Falls

Fri, 09/19/2014 - 7:48am

By: Martin Olszynski

Case commented on: Council of the Innu of Ekuanitshit v. Canada (Attorney General), 2014 FCA 189 (CanLII)

At least three times in the course of the past year, an environmental assessment (EA) panel convened under the Canadian Environmental Assessment Act 2012, SC 2012, c 19 (CEAA, 2012) has concluded that a project is likely to result in significant adverse environmental effects: Shell’s Jackpine Mine Expansion, Taseko’s New Prosperity Mine, and Enbridge’s Northern Gateway Pipeline.  In the case of both Jackpine and Northern Gateway, the federal Cabinet determined that these effects were “justified in the circumstances,” but not so for New Prosperity. In none of these instances, however, did the relevant “Decision Statement” pursuant to section 54 of CEAA, 2012 contain any explanation or reasons for Cabinet’s decision. The Federal Court of Appeal’s recent decision in Council of the Innu suggests that this approach is wrong. This litigation involved the Lower Churchill Hydroelectric Project proposed by Nalcor in Newfoundland. This project was reviewed under the previous CEAA regime but the relevant provisions are virtually unchanged. Like the three EAs referred to above, the panel concluded that the project was likely to result in significant adverse environmental effects. Unlike the three projects referred to above, however, the government did provide a detailed explanation for its determination that the significant adverse environmental effects were justified in the circumstances. The Council challenged this determination (the Council also challenged the sufficiency of Aboriginal consultation; this post focuses only on the justification issue). 

The Council’s primary argument was that the project as proposed and assessed involved two plants, a larger Gull Island plant and a smaller Muskrat Falls plant, but that at the time of Cabinet’s decision-making a construction date for the Gull Island plant had yet to be confirmed, such that it was unreasonable for Cabinet to rely on the benefits of both plants when making its decision about justification. The Federal Court of Appeal ultimately disagreed (see para 58), but not without first setting out a framework for reviewing a “justified in the circumstances” determination. Beginning with the standard of review, the Court endorsed the trial judge’s approach:

[40] …the Court will only intervene with the [Governor in Council’s] and Responsible Ministers’ decisions under subsections 37(1.1) and 37(1) if it finds that: 1) the CEAA statutory process was not properly followed before the section 37 decisions were made; 2) the GIC or Responsible Ministers’ decisions were taken without regard for the purpose of the CEAA; or 3) the GIC or Responsible Ministers’ decisions had no reasonable basis in fact; which is tantamount to an absence of good faith.

The vast majority of CEAA litigation has focused on the first criterion, with applicants alleging various deficiencies with the EA process. In fact, to the best of my knowledge the Council of Innu decision is the first to challenge the legality of the GiC’s decision directly. And while the parameters of review here (the second and third criteria above) are deferential to be sure, it is equally clear that they require something by way of explanation.  Otherwise, it is simply impossible to carry out what the Court of Appeal stated was its duty (at para 44): “a reviewing court must ensure that the exercise of power delegated by Parliament remains within the bounds established by the statutory scheme” (essentially the same approach I argued for here). Turning to the GiC order with respect to Lower Churchill, the Court noted (at para 53) that Cabinet “determined, after consulting the Joint Review Panel Report as well as several government studies, that [translation] ‘the significant energy, economic, socio-economic and environmental benefits outweigh the negative environmental impacts of the Project identified in the Panel’s Report.’” The Court then addressed the Council’s argument with respect to the Gull Island Plant in a passage that suggests that, notwithstanding its subjective and policy-laden nature, the justification determination must be able to withstand at least some scrutiny:

[54] I share the appellant’s view that the abandonment of the Gull Island plant, if this were proven to be true, would raise serious questions about the validity of the environmental assessment and the impugned decisions. The Project authorized by the Governor in Council and responsible authorities following the balancing exercise imposed by section 37 of the CEAA included the Muskrat Falls plant as well as the Gull Island plant… If Nalcor were to forego construction of the larger of the two plants assessed (Gull Island), or if there was an unreasonable delay in its construction, the balancing exercise carried out for one of the Report’s findings would be necessarily compromised.

As noted above, the Court of Appeal ultimately concluded that the Council failed to prove that the Gull Island plant had been abandoned or that it had been unreasonably delayed (at para 57), but the above framework and its application to this case leave little doubt that the federal government’s current practice with respect to justification is insufficient.

Nor would it seem sufficient for the government to simply rely on the justification occasionally provided by panels as in the case of Jackpine, which was a joint review panel with the Alberta Energy Regulator and which concluded that that project was in the public interest, or Northern Gateway. As a starting point and consistent with the Federal Court’s recent decision in Greenpeace Canada v Canada (Attorney General), 2014 FC 463 (CanLII), such panels lack democratic legitimacy.  Another reason, and something that I have noted before, is that there often exists a yawning gap between panel recommendations and the conditions that the government ultimately imposes on proponents (see e.g. the recommendations with respect to a Traditional Land Use (TLU) management framework under the Lower Athabasca Regional Plan in the context of Shell’s Jackpine project). Practically, this means that there is often a real difference between the “balance” reached by panels and that struck by the government.

Judicial Dissent over Priorities in Civil Justice: Queue-Jumping and the Commercial List

Tue, 09/16/2014 - 10:00am

By: Sarah Burton

PDF Version: Judicial Dissent over Priorities in Civil Justice: Queue-Jumping and the Commercial List

Cases Considered: Lustre Studio Inc. v West Edmonton Mall Property Inc, 2014 ABQB 525

In Lustre Studio Inc. v West Edmonton Mall Property Inc, 2014 ABQB 525, the Honourable Mr. Justice B.R. Burrows provided a candid window into judicial frustrations with access to justice in Alberta.  In pointed words, he expressed dissatisfaction with the courts’ willingness to prioritize and accommodate commercial cases through mechanisms unavailable in family and non-commercial matters. While Justice Burrows clearly criticizes this preferential treatment, he also expresses resignation in quelling the tide. This decision implicitly questions the priorities of our justice system and the preference given to commercial matters over non-commercial cases, even when they urgently require the court’s attention. Practically speaking, Justice Burrows may be correct in stating that expanded accommodations for commercial cases are here to stay. If so, this innovative project should be harnessed to create equally effective mechanisms for family and other non-commercial cases.

Facts

On August 22, 2014, Justice Burrows was completing a week on Edmonton Chambers Duty (sitting in Family, Regular and Special Chambers). Many of the matters he heard in Family and Regular Chambers were complicated and required a Special Chambers hearing. Unsurprisingly, however, there was a significant queue for Special Chambers dates stretching until the winter and/or spring of 2015.

That same day, Justice Burrows received a request from counsel for Lustre Studio Inc. (“Lustre”) to have an interim injunction application heard as part of the Commercial/Duty Justice Initiative at a full day hearing on September 12, 2014. The injunction revolved around a commercial lease dispute between Lustre and its landlord West Edmonton Mall Property Inc. (“West Ed”), wherein West Ed invoked a lease provision requiring Lustre to relocate by January 16, 2015, which Lustre sought to avoid.

Lustre’s effort to have its case heard in September 2014 (rather than early 2015) is permitted by the Commercial/Duty Justice Initiative (the “Commercial List”). The Commercial List is a 2010 initiative of the Court of Queen’s Bench (Notice to the Profession #2010-08) that creates a separate queue for commercial cases to access specialized judges on an expedited basis. Practically speaking, if a case falls within the Commercial List mandate, it can be heard by a judge from the Commercial Practice Group within weeks, not months. While it began as a Calgary/Edmonton project to accommodate bankruptcy, insolvency and related matters, the Commercial List was expanded in a Notice to the Profession issued by the Chief Justice and Associate Chief Justice of the ABQB (NP#2014-04) in April 2014 to include non-bankruptcy commercial matters such as:

  • Mareva Injunctions;
  • Anton Pillar Orders;
  • Third-Party Production (Norwich) Orders;
  • Remedies under Business Corporations or Securities legislation; and
  • Other matters permitted by a Commercial Duty Judge sitting in Commercial Appearance Court or a Co-Chair of the Commercial Practice Group or her designate (at paras 5, 6).

Counsel for Lustre relied on the last of these headings.  In its letter to Justice Burrows, Lustre did not allege that its matter was urgent, as urgency is not a requirement of the expanded Commercial List initiative (at para 8).

Reasons

Justice Burrows granted Lustre’s request (at para 12). The expanded Commercial List initiative did not provide guidance for his exercise of discretion. Since it fit within the mandate and nothing else was scheduled on that day, he had no reason to refuse the request.

Justice Burrows explained, however, that he was granting the request despite his fundamental disagreement with the expanded Commercial List initiative. While it is uncharacteristic for a judge to highlight his disagreement with a law, Justice Burrows was compelled to make this comment because the law in question was a judicial policy implemented by the Alberta judiciary. As such, he wished to make his personal disagreement with this law crystal clear. In his words:

[15] … My oath requires me to apply the law even when I am of the view that the law is not what it should be. This is a situation where I am obliged to apply “law” with which I fundamentally disagree.

[16] Ordinarily, it would not be relevant or even appropriate for a judge to point out where he does not agree with the law he is obliged to apply.  In my view that is not the case where, as here, the judge is a member of the policy setting body which adopted the law in question and might otherwise be thought to have agreed with the law as adopted.

Justice Burrows’ objection was not based in opposition to Lustre’s argument, injunction applications, lease disputes, or commercial matters generally. Rather, he was motivated by the inequity of dedicating scarce judicial resources to prioritize commercial matters over family and non-commercial civil actions. The Commercial List plainly prioritizes non-urgent commercial cases over urgent family and non-commercial civil matters. Thus, while bound to apply the law, Justice Burrows expressed his palpable distaste for it. He concluded:

[17] The Court clearly requires a triage system to deal with matters which have an element of urgency.  In this Court there is no formal triage system.  A family and non-commercial civil litigant who feels their matter requires priority judicial attention can do little more than hope that a judge will agree and make some ad hoc arrangement for the early hearing of the application.  As noted, pursuant to the Notice to the Profession, commercial business matters are simply given automatic priority assuming there is free time on the schedule of the commercial duty judge.  In Edmonton at least, such free time occurs frequently.

[18] I have on several occasions made my view on this subject known to my colleagues.  I have been unable to prevent the adoption of this Notice to the Profession.  I record here that, though I am bound to apply it, I believe it institutes an extremely ill-advised policy.

Commentary

The Priority of Commercial List

Commercial litigators in Calgary and Edmonton often praise the efficiency, specialization, and practicality of the Commercial List. As a former commercial litigator, I have some familiarity with the Commercial List where matters are often undeniably urgent and the stakes are high. The Commercial List itself is not inherently bad, nor does it impede access to justice. It is an intelligent response to a problem facing commercial litigants who need to access the courts on time sensitive matters. The Commercial List has undeniably advanced access to justice for many commercial parties who (for example) face immediate and irreparable harm from a creditor-induced bankruptcy application, or a permanent injunction that could destroy a business and put many people out of work.

The problem with the Commercial List arises from its inequity in relation to non-commercial cases. While certain commercial litigants are able to access a formal queue-jumping system, as Justice Burrows stated “[a] family and non-commercial civil litigant who feels their matter requires priority judicial attention can do little more than hope that a judge will agree and make some ad hoc arrangement for the early hearing of the application” (at para 17).

As matters stand, the expanded Commercial List effectively creates a two-tiered system whereby non-urgent commercial matters are blatantly prioritized over urgent family and non-commercial civil cases. This distinction is unfounded and unacceptable.

I would not, however, suggest that we abolish the Commercial List to rebalance the current inequity. It is an example of a positive and practically based solution to the problem of accessing courts.  Therefore, instead of attacking the program, we ought to use it as an example. The same innovative thinking used to develop the Commercial List (and the resources needed to implement it) would be well-utilized developing a similar program for family and non-commercial litigants in urgent cases. This would rebalance the current inequity while promoting innovative solutions to make courts more accessible.

The Emergence of a Trend?

On a concluding note, the timing of Justice Burrows’ comments also merits discussion. Earlier this month, I blogged on R v Smart, 2014 ABPC 175, where the Honourable Assistant Chief Judge Anderson stayed proceedings against three accused persons who could not afford counsel, but did not qualify for Legal Aid. Like Lustre, Smart contained considerable commentary on the barriers to justice faced by parties who fail to fit within a particular initiative (in that case, Legal Aid). While couched in different terms, both decisions expressed frustration with the justice system’s failure to adequately serve the people who use it.

Strikingly, despite their stated views on increasing access to justice, in both cases the adjudicators ultimately sacrificed civil and family law matters in favor of cases with a greater priority. In Lustre, Justice Burrows was ultimately required to accommodate the applicant’s request to the detriment of non-commercial civil and family law cases. In Smart, Judge Anderson leveraged the constitutional priority given to criminal cases to demand that the accused persons be appointed Legal Aid counsel. As discussed in my earlier blog, this decision will likely come at the expense of civil and family litigants seeking Legal Aid.

Lustre and Smart were released approximately two weeks apart. It is likely premature to label two cases a “trend”, but they are consistent with recent reports on access to civil justice, and this recent judicial commentary does provide an invaluable perspective on how barriers to justice are directly encountered and perceived in the courtroom. These decisions also highlight the fact that barriers to justice are often systemic. In their decisions, Judge Anderson and Justice Burrows both clearly wanted to increase access to justice. Their decisions, however, were ultimately compelled by the application of a triage system that prioritizes some cases over others. Unfortunately, when there aren’t enough resources to go around, family and non-commercial civil matters are continually left in the cold. Clearly, this inequity cannot be remedied on a case-by-case basis in the courtroom. Policies and (before that) perspectives on the importance of civil justice must change if we want to level the playing field.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

Federal Court to Syncrude: Climate Change is a Real, Measured Evil, Whose Harm has been Well Documented

Mon, 09/15/2014 - 9:29am

By: Martin Olszynski

Case commented on: Syncrude Canada Ltd. v Attorney General of Canada, 2014 FC 776

“The fall term in the 1997-1998 academic year,” wrote Professor David Beatty, “was a constitutional law teacher’s dream.” Professor Beatty was referring to the release of two Supreme Court of Canada decisions that touched some of the “most politically charged issues” of the day and which “together raised almost every important issue in constitutional law” (one of which was R v. Hydro Quebec, [1997] 3 SCR 213, 1997 CanLII 318 (SCC), central to the Syncrude decision being commented on here; see David Beatty, “Canadian Constitutional Law in a Nutshell” (1998) 36(3) Alta L Rev 605). As it turns out, the summer of 2014 has shaped up to be an environmental law teacher’s dream.  In May, the Federal Court released its decision in Greenpeace Canada v Canada (Attorney General), 2014 FC 463 (CanLII), a decision that I have suggested represents a major development in Canadian environmental assessment law. Then in August, the Federal Court handed down its judgment in Syncrude, which my colleague Professor Nigel Bankes has observed is the “first case in which a party has challenged the constitutional validity of any federal greenhouse gas regulations.”  This post focuses on that very issue; Professor Shaun Fluker has also written a post on the decision, focusing on the administrative law issues.

The Constitutional Question

Subsection 5(2) of the Renewable Fuel Regulations, SOR/2010/189 (RFR), being regulations made pursuant to subsection 140(1) of the Canadian Environmental Protection Act, 1999, SC 1999 c 33 (CEPA), requires that diesel fuel produced, imported or sold in Canada must contain renewable fuel of at least 2% by volume.  Compliance can be achieved by either blending diesel with biodiesel, which is made from either biological waste matter or feed stocks, or by purchasing compliance units from those who have more than 2% renewable fuel in their diesel fuel.

Canada argued that subsection 5(2) “is in pith and substance a legitimate use of the federal criminal law power to suppress the evil of air pollution by mandating a 2% renewable fuel content in diesel fuel produced” (at para 12). For its part, Syncrude argued that the provision’s “dominant purpose and effect…is to regulate non-renewable resources and promote the economic benefits of protecting the environment,” and in particular “to create a demand for biofuels in the Canadian market place,” such that any prohibition of harm is merely ancillary (ibid).

Justice Zinn first set out the applicable framework as recently set out in Québec (Procureur Général) v Canada Procureur (Procureur Général), 2010 SCC 61 (CanLII) [Re: Assisted Human Reproduction]. The first step is to determine the dominant matter – the pith and substance – of the impugned provisions, a contextual analysis that requires asking “[w]hat in fact does the law do and why?” (at para 16, citing para 22 of Re: Assisted Human Reproduction). The second step is to determine which of the heads of power such matter falls under.

Justice Zinn then began by considering the purpose of the RFR and, because it is subordinate legislation, the CEPA as well. Referring to the preamble of both, Justice Zinn observed that CEPA is designed “to address environmental degradation, protect the environment and human health, and place the cost and responsibility of pollution on the polluter”, while the RFR were considered by the Governor in Council (GiC) as contributing to “the prevention of, or reduction in, air pollution…” (at para 17).  Following the teaching in Bristol-Myers Squibb Co v Canada (Attorney General), 2005 SCC 26 (CanLII), Justice Zinn also considered the various Regulatory Impact Analysis Statements (RIAS) related to the RFR, which in his view confirmed that the driving concern was the reduction of greenhouse gas (GHG) emissions (at para 21).  Readers interested in the history and evolution of the federal government’s position on climate change should check out paragraphs 22 – 28 of the judgment. Justice Zinn himself summarized those documents as follows:

[30] …The RIAS for both the RFR and its amendment…make clear that GHG emissions pose a significant, enduring effect on the environment, have high global warming potentials, and can directly affect the health of Canadians.  The RIASs also explain that renewable fuels have been shown to make a significant contribution to lowering GHG emissions on a life-cycle basis.  While the provinces currently have regulations imposing renewable fuels requirements, Parliament was of the view that federal regulation could contribute above and beyond the provincial contributions and would fill gaps and address inconsistencies in provincial legislation.

While Justice Zinn recognized that the RFR were “also intended to increase the demand for renewable fuels and develop new market opportunities” (at para 31), it was equally clear that these economic effects were part of a larger strategy, the purpose of which was to reduce GHG emissions: “Creating a demand for renewable fuels was…a necessary part of the overall strategy to reduce GHG emissions, but it was not the dominant purpose.  The reason the government wanted to create a demand for the fuels was to make a greater contribution to the long term lowering of GHG emissions” (at para 35).

Consequently, Justice Zinn had no difficulty concluding that “the dominant purpose of the RFR was “to make a significant contribution to the reduction of air pollution, in the form of reducing GHG emissions” (at para 39). Turning to the regulations’ effect, the bulk of this part of the decision was aimed at rejecting Syncrude’s efforts to have the Court assess the efficacy of the RFR. However, Justice Zinn was also of the view that even if this was part of the test, Syncrude “has failed to present convincing evidence to show that the blending of renewable fuels would not ‘make a significant contribution to the prevention of, or reduction in, air pollution’” (at para 45; see also para 49).

Having determined that the pith and substance of the RFR is the reduction of GHG emissions and potentially other emissions (at para 54), Justice Zinn turned to classification.  The federal criminal law power requires that there be (1) a prohibition; (2) backed by a penalty; (3) with a criminal law purpose (Reference re Firearms Act (Can), 2000 SCC 31 (CanLII)). There was apparently no dispute with respect to the first two criteria; the issue was whether the RFR were enacted with a valid criminal law purpose (at paras 58, 59).

The Minister relied on R. v. Hydro Quebec, the landmark but also controversial Supreme Court ruling that grounded CEPA’s toxic substances regime in the criminal law power. In that case, both the majority and minority agreed that Parliament’s use of the criminal law power was not limited to protecting public health; “the protection of the environment is itself a legitimate basis for criminal legislation” (Hydro Quebec, at para 43). Where the parties disagreed, and where the majority ultimately prevailed, was whether the impugned regime was more regulatory than prohibitory in nature. Writing for the majority, Justice La Forest held that environmental issues did not always lend themselves to blunt prohibitions but that this should not preclude Parliament from relying on the criminal law power, especially where the alternative is the seemingly more drastic reliance (from a federalism perspective) on its residual power to pass laws for peace, order and good government (POGG).

Applying this reasoning to the RFR, Justice Zinn acknowledged that they appeared “more regulatory in nature than prohibitory.  However, like the majority in Hydro, I am of the view that this particular evil – GHG emissions by combustion of fossil fuels – is not well addressed by specific [sic] prohibitions.  For example, much of society runs on fossil fuels and Parliament should not be expected to prohibit the use of fossil fuels entirely in order to meet progressive goals of GHG emission reduction” (at para 67).

Where the decision gets really interesting, in my view, is around paragraph 78. Here, and in contrast to the public messaging from industry and industry groups such as the Canadian Association of Petroleum Producers (i.e. that they support action on climate change), Syncrude argued that “the production and consumption of petroleum fuels is not dangerous and does not pose a risk to human health or safety,” and that “there is no evil to be suppressed.” To this, Justice Zinn responded that “[t]he evil of global climate change and the apprehension of harm resulting from the enabling of climate change through the combustion of fossil fuels has been widely discussed and debated by leaders on the international stage.  Contrary to Syncrude’s submission, this is a real, measured evil, and the harm has been well documented” (at para 83).

Discussion

Several years ago at the annual “CBA-DoJ Day” in Ottawa (an event organized by the National Energy, Environment and Resources Law section of the Canadian Bar Association), someone asked counsel from Environment Canada if they thought that the toxic substances provisions of CEPA – the provisions upheld under the criminal law power in R. v. Hydro Quebec – would pass muster if they were to be challenged now, in light especially of the fact that there are now thousands of regulated substances under that scheme. If I recall correctly, the response was positive if somewhat tepid. Justice Zinn’s decision would seem to be put any doubt to rest, while at the same time swinging the criminal law door open for all manner of schemes to control GHG emissions, including cap and trade. And while it is pretty clear that this door is not one that the current federal government will be walking through, a future government just might.

Syncrude v Canada: Where is the gatekeeper when you need one?

Fri, 09/12/2014 - 10:00am

By: Shaun Fluker

PDF Version: Syncrude v Canada: Where is the gatekeeper when you need one?

Case commented on: Syncrude Canada Ltd v Attorney General of Canada, 2014 FC 776

This post continues on from the introductory comment posted by Nigel Bankes on September 11, 2014 (here) concerning this case, and discusses the administrative law aspects in Justice Zinn’s decision. Briefly put, Syncrude challenges the validity of the Renewable Fuels Regulations, SOR/2010-109 enacted pursuant to section 140 of the Canadian Environmental Protection Act, 1999, SC 1999, c 33 [CEPA]. Section 139 of CEPA together with the Renewable Fuels Regulations require diesel fuel produced, imported or sold in Canada to contain renewable fuel of at least 2% by volume.  Syncrude produces diesel fuel, and is thus subject to this requirement unless it can successfully argue the Renewable Fuels Regulations are ultra vires the authority of the  Governor in Council or that there is some other legal defect in how the rules have been administered against it. My comment focuses on two points in the decision, namely: (1) are the Renewable Fuels Regulations unlawful because they do not conform to the regulation making powers of the Governor in Council set out in section 140 of CEPA?; and (2) did the Minister err in law by failing to afford Syncrude procedural fairness in administering the regulations?

I would say the resolution of these points by Zinn J. represents a fairly straightforward application of settled law, and I can accordingly describe the essence of the reasoning in short order.  The law on the first issue was recently summarized by the Supreme Court of Canada in Katz Group Canada v Ontario (Health and Long-Term Care), 2013 SCC 64 at paras 24-28. An issue over the vires of a regulation involves the following considerations: (1) is the impugned regulation consistent with the objective of its parent statute – in order to demonstrate invalidity a person must establish that the regulation is not consistent with such objective or that it addresses a matter which is not set out in the regulation-making provision of the parent statute; (2) there is a presumption of validity such that the onus or burden is on the challenger to demonstrate that the regulation is ultra vires – so where possible a regulation will be read in a ‘broad and purposive’ manner to be consistent with its parent statute; (3) the inquiry into the vires of a regulation does not involve assessing the policy merits of the regulation, nor does the reviewing court assess whether the regulation will successfully meet its objective. There is also, of course, always the prospect that a regulation is unlawful because it has been enacted for an improper purpose that amounts to an abuse of discretion.

After citing the Katz decision, Justice Zinn considers various arguments put forward by Syncrude to support its position that the Renewable Fuels Regulations are invalid because they fail to conform with CEPA. Zinn J. rejects all of these arguments, and I will comment on these in turn.

Syncrude argues the Governor in Council erred by failing to form the opinion that the Renewable Fuels Regulations could make a significant contribution to the prevention of, or reduction in, air pollution before enacting the regulations – an opinion which section 140(2) of CEPA requires it to form. Here Justice Zinn references the preamble to the regulations, which states such an opinion was formed and he applies the presumption of validity noted above to assert that the validity – or in my words the bona fides – of this opinion as set out in the regulations is not open to question in this case without tangible evidence from Syncrude that such opinion was not actually formed (at paras 114 – 118).

Syncrude argues a Cabinet Directive on strategic environmental assessments concerning policy proposals imposes a mandatory obligation on a Minister to perform such an assessment before enacting the Renewable Fuels Regulations. This argument required Syncrude to establish that the Cabinet Directive is law rather than policy, which it fails to do (at paras 119-123). Justice Zinn is rather short on reasoning here, and perhaps little was made of this in argument. However, the distinction between legislative instruments which are binding and policy instruments which are not binding is a complex area, and resolving these questions is typically more involved than what is set out in this case. A good summary of the applicable law on this point can be found in John Mark Keyes, Executive Legislation, 2d ed (Lexis Nexis, 2010) at 19 – 63.

Syncrude argues the Renewable Fuels Regulations do not accord with the purpose and objective of CEPA because they do not protect the ‘environment’ as defined in CEPA to mean air, land and water. In this regard, Syncrude emphasizes the conjunctive aspect of this definition and asserts that the record did not include material to demonstrate the regulation was intended to protect land and, moreover, that the renewable fuel requirement would actually result in negative impacts to these components. Again citing the Katz decision, Justice Zinn rules that a vires inquiry cannot go into the merits of the policy underlying the regulation save for those egregious cases where abuse of discretion is evident. And moreover, he lists several reasons for rejecting the conjunctive interpretation of ‘environment’ offered by Syncrude (at paras 124 – 137).

The second issue concerns whether the Minister failed to afford Syncrude procedural fairness in rejecting Syncrude’s request for a review of the Renewable Fuels Regulations before they were enacted. Syncrude argues that before rejecting such request, the Minister was required to consult with Syncrude on the issue and also that the Minister provide Syncrude with reasons for its decision not to initiate the review. This argument by Syncrude is based on the Minister’s power under sections 332 and 333 of CEPA to establish a board of review – on the request of a person who objects to a proposed regulation – to inquire into the nature and extent of the danger posed by the substance in respect of which the regulation is proposed. The legal question here is whether the Minister owed Syncrude a duty of procedural fairness under the common law, since the statute is silent on these questions of consultation and reasons.

Justice Zinn has little trouble ruling the Minister has no such duty of procedural fairness towards Syncrude or any other person for that matter in deciding whether or not to establish a board of review (at paras 144 – 161).  The law on this point is well settled: The common law duty of procedural fairness does not apply to decisions of a legislative or general nature. Syncrude argues the decision is specific and directed at it, however Zinn J. points to the wording in CEPA that any such review would address general matters pertaining to the substance in question under the regulation rather than the specific circumstances of a person objecting to the regulation.

I will end this comment by returning to the title – where is the gatekeeper when you need one? The constitutional law aspects of this decision aside, which I have not considered here, it seems to me there was little prospect of success on these administrative law arguments by Syncrude.  One can’t help but read this judgment and contrast it with the many leave to appeal decisions of the Alberta courts on applications involving Alberta’s energy regulatory boards, which seem to hold the applicants to an unbearably high burden of showing not just the prospect of success but more than a reasonable probability of success in their arguments, before even getting the chance to argue their case on the merits!

The Federal Renewable Fuels Regulations Survive an Aggressive and Comprehensive Challenge from Syncrude

Thu, 09/11/2014 - 10:00am

By: Nigel Bankes

PDF Version: The Federal Renewable Fuels Regulations Survive an Aggressive and Comprehensive Challenge from Syncrude

Case commented on: Syncrude Canada Ltd. v Attorney General of Canada, 2014 FC 776

In the dog days of summer (August 6, 2014) Justice Russel Zinn of the Federal Court of Canada handed down his judgement in a case in which Syncrude sought to challenge the validity of the federal Renewable Fuels Regulations, SOR/2010/189 (RFR) on both constitutional and administrative law grounds. The judgment seems to have passed almost without comment in the media. The RFR require that diesel fuel produced, imported or sold in Canada must contain renewable fuel of at least 2% by volume. This requirement can be met by blending diesel with biodiesel (although this can be challenging at cold temperatures). Failure to comply with the RFR is an offence although a regulated entity can achieve compliance by purchasing compliance units from other regulated entities who have exceeded their own compliance targets. See the judgement at para 4.

It is a pity that the decision did not attract greater media attention because the case is politically important and also legally significant on a number of fronts. For example, this is the first case in which a party has challenged the constitutional validity of any federal greenhouse gas regulations. The decision gives strong support to the federal government’s reliance on the criminal law power to legislate in this area. The case also offers useful guidance at the administrative law level as to the preconditions that the Governor in Council must meet before making regulations under the Canadian Environmental Protection Act, 1999, SC 1999, c.33. The decision confirms that the courts will not interfere with the manner in which the executive goes about meeting its policy objectives even if the efficacy of the measure (either absolutely or comparatively) may be open to question. And there are other interesting tidbits in the case as well. For example, there is a useful discussion of the Cabinet Directive on Strategic Environmental Assessments and on the distinction between a Cabinet Directive and a regulation made by Order in Council. We hope to post individual comments on some of the specific legal issues over the coming week(s).

Finally, it is interesting to consider the greenhouse gas politics of this decision. This set of federal regulations represents a tiny, tiny step towards meeting Canada’s post-Kyoto diluted Copenhagen challenge of minus 17% of 2005 emission levels by 2020. I am sure that compliance with the regulations will increase operating costs but I am equally sure that this increase is nothing more than a drop in the bucket for Syncrude. This litigation therefore suggests that at least the sector of big oil represented by the Syncrude interests will fight federal greenhouse gas regulations in all of its forms and that it will fight them hard. There was no stone left unturned in this litigation. Counsel for Syncrude pursued every possible avenue no matter how small the chance of success or creative the argument. Big carbon may be just like big tobacco in protecting its turf – one of several similarities that my colleagues Martin Olszynski and Sharon Mascher hope to explore more fully in a forthcoming research project.

Big carbon might want to think more carefully about the politics of this. So who stands behind Syncrude? Who should the press be asking questions of? Answer: (37%) Canadian Oil Sands Partnership, (9%) SINOPEC, (25%) Imperial Oil Resources, (5%) Mocal Energy, (5%) Murphy Oil Company, (7%) Nexen Oil Sands Partnership, (12%) Suncor Energy Ventures Partnership. Perhaps Imperial Oil would be a good place to start given their important global carbon position.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

The Vicious Spiral of Self-Representation in Family Law Proceedings

Wed, 09/10/2014 - 10:00am

Written by: John-Paul Boyd

PDF Version: The Vicious Spiral of Self-Representation in Family Law Proceedings

A lot of good research on litigants without counsel has been published in the last three years, most notably, in my view, Julie Macfarlane‘s Identifying and Meeting the Needs of Self-represented Litigants, a trio of papers published by the Canadian Research Institute on Law and the Family on the views of Alberta judges and family law lawyers, and a report by the Canadian Research Institute on Law and the Family with professors Nicholas Bala and Rachel Birnbaum (in press) on the results of a national survey of judges and lawyers. Although this research doesn’t necessarily label it as such, I’ve noticed that there’s a bit of a slippery slope effect to litigating without counsel, in which the decision to self-represent, whether a choice was involved or not, seems to trigger a cascade of adverse effects that ultimately result in litigants without counsel achieving worse results in every major area of family law than would have been achieved with counsel.

The judges and lawyers surveyed firstly said that by and large litigants without counsel have unrealistically high expectations of the outcomes they are likely to achieve. (In the Alberta research, three-quarters of judges and almost 90% of lawyers said that litigants without counsel “always” or “usually” have unrealistic expectations of outcome; almost half of judges and lawyers outside Alberta said the same thing in the national survey, and 30% said that litigants without counsel “sometimes” have unrealistic expectations of outcome.)

Perhaps as a result of their overly optimistic expectations, litigants without counsel are more likely to go to trial than settle. (In Alberta, 87% of judges and 89% of lawyers said that settlement without trial or before the end of trial is “less likely” or “much less likely” if one party is self-represented. In the national study, almost 60% of judges and lawyers said that settlement is “less likely” or “much less likely”. The national survey also showed that 70% of respondents from Alberta and 55% of respondents from the rest of Canada believe that litigants without counsel are more likely to take unreasonable positions based on principle.)

When litigants without counsel get to trial, problems arise as a result of their unfamiliarity with the law and court processes. (In the Alberta study, a range of 85 to 100% of judges said that additional challenges “always” or “usually” arise in cases involving a litigant without counsel because of their unfamiliarity with the applicable legislation, the rules of evidence, the rules of court and hearing and trial processes. A range of 70 to 77% of judges and lawyers outside of Alberta said the same thing in the national survey.)

At the end of the day, litigants without counsel achieve worse results than litigants with counsel. (In the national research, a range of 51 to 55% of judges and lawyers in Alberta thought that litigants without counsel achieved worse results on child support and spousal support, parenting arrangements and the division of property; 32 to 44% of judges and lawyers in the rest of Canada felt the same way. About 18% of judges and lawyers said that there is no difference in the results achieved by self-represented litigants on support issues, about 20% said that there is no difference in the results on support issues and about 13% said that there is no difference in the results for property division.)

(However, despite their difficulties with court and court processes, litigants without counsel are generally treated well by judges. In The Rise of Self-representation in Canada’s Family Courts (summary), a study which included a survey of Ontario litigants, Nicholas Bala, Rachel Birnbaum and Lorne Bertrand found that 14% of litigants without counsel believed that they were treated “very well” by the bench, 39% thought the way they were treated was “good” and 13% thought they were “not treated well at all.” According to the national survey, 97% of the lawyers from Alberta and 74% of the lawyers from the rest of Canada believe litigants without counsel receive “very fair” or “fair” treatment from the bench.)

It appears from the research that litigants without counsel find themselves caught in a vicious spiral. Self-represented litigants generally have unrealistically high expectations for the outcome of their cases, which reduces the likelihood that their cases will be resolved without trial. When they do proceed to trial, their lack of knowledge of the governing legislation, the rules of evidence, the rules of court and court processes frequently causes additional problems and doubtless increases the length of trials and the number of adjournments, and, when their trials do complete, self-represented parties usually achieve worse results than they would have with counsel.

I’ll provide some thoughts on ways of interrupting the slippery slope in a future post; in the meantime, the results of the research on the attitudes and outcomes of litigants without counsel should be factored into the family justice reform work being undertaken across the country. Mandatory mediation and other diversionary steps strike me as an obvious contribution to a solution.

This comment was originally posted on Access to Justice in Canada.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

The Quiet Decline of Canada’s IPO Markets

Tue, 09/09/2014 - 10:00am

By: Bryce Tingle

PDF Version: The Quiet Decline of Canada’s IPO Markets

The Toronto Stock Exchange’s parent company has been travelling the country raising the profile of its new venture, TSX Private Markets.  At the same time, Canada’s securities commissions are engaged in the most comprehensive overhaul of the private placement regime in more than a decade.  In Ontario, in particular, this would reverse the increasingly restrictive trends of previous reforms and liberalize its private capital markets.

This is a curious state of affairs. The TSX is chipping away at the incentives for a company to go public and the Ontario Securities Commission (OSC) is making it easier for companies to raise money outside of its regulatory “gold standard”: the public company prospectus system.  What is going on?

Canada’s public markets are failing to do their job of supplying capital to innovative businesses in this country.  Canada is uniquely dependent on its public markets.  We have the largest number of public companies relative to our population in the world – more than double the next highest country (and four times higher than the United States).  But our Initial Public Offering (IPO) market has been collapsing for over a decade.

Research published in the most recent issue of the Canadian Business Law Journal by Ari Pandes, Michael Robinson and myself, sets out, for the first time that we are aware of, an analysis of the rate operating businesses have been going public in Canada since statistics became readily available in 1993.  Between 1993 and 2000 there were an average of 42.6 IPOs on the TSX each year; following 2000 the average was 18.2 IPOs a year – less than half.  Even when the markets were awash in liquidity and commodity prices climbed sharply, TSX IPOs in the best years of this century never exceeded the average rate of the previous decade and only amounted to 40% of its best year.

The decline becomes even more visible when the inflation-adjusted proceeds of IPOs over the past two decades are examined.  Both in absolute terms and as a percentage of Gross Domestic Product (GDP), the IPO market of the 1990s was in significantly better health than at any time in the years since then.  The best year this century for IPOs was 2010, which generated proceeds equal to 0.28% of GDP – or just over half of the proceeds generated during any one of several years in the 1990s.

Reference to the dot-com boom and bust, and then the 2008 financial crisis, provide little assistance in explaining these statistics.  The Canadian market is largely resource-driven, not particularly oriented towards technology stocks.  Low commodity prices in the second half of the 1990s likely explains why this country’s IPO rates actually declined, even as the United States IPO market developed a bubble.  Similarly, Canada’s IPO market was only comparatively briefly impacted by the 2008 crisis, with 2010 being the best year for IPOs in the past decade.

The United States has convened congressional hearings (see here and here) and blue-ribbon panels (see here, here and here) to look into the causes of the decline in public companies in that country.  Canada has done nothing.  But the fact that we are experiencing the same kind of decline as the United States makes us an important part of any analysis of the problem.  The most common explanation in America for the decline in IPOs is the cost of complying with Sarbanes-Oxley era regulation – but Canada did not import most of this regulation and avoided, in particular, the most expensive requirements such as the auditing of financial controls.

Other common American explanations for the decline of IPOs, such as America’s toxic litigation climate (see here and here), changes in the benefits conferred by scale in the high-tech industry, or alterations to the market structure affecting investment banks (see here and here), also run afoul of the Canadian counterfactual.

We are left with the simplest of all explanations: the public markets have become much less congenial to business.  Spend any time in Canada’s boardrooms these days and you will hear the same message.  Whereas there were once incentives to management teams to take their companies public, there are now incentives to avoid the public markets if at all possible.  The result is a climbing cost of capital, less transparency and fewer investment opportunities for a Canadian middle class increasingly responsible for its own retirement.

Professor Edward J. Waitzer, has recently argued for a comprehensive review of Canada’s capital markets by our securities regulators to ensure we are working on our most urgent priorities as a country.  I second the motion.

An earlier version of this comment appeared in the National Post.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

R v Navales and Reasonable Suspicion

Fri, 09/05/2014 - 10:00am

By: Shaun Leochko

PDF Version: R v Navales and Reasonable Suspicion

Cases Commented On: R v Navales, 2014 ABCA 70; R v Canlas, 2014 ABCA 160; R v Ng, 2014 ABPC 62; R v Tosczak, 2014 ABQB 86

The engagement of section 8 and section 9 of the Canadian Charter of Rights and Freedoms (the Charter) in the drug sniffer dog cases has captured the interest of civil libertarians and law enforcement for what is required for a “reasonable suspicion.”  The 2013 Supreme Court decisions of R v Chehil, 2013 SCC 49, and R v MacKenzie, 2013 SCC 50 effectively lowered what would be required of police officers to form the reasonable suspicion necessary to conduct a “sniff” search. This resulted from the Supreme Court allowing an officer’s training and experience, in the totality of the circumstances, to form the objective requirement necessary to the forming of reasonable suspicion.  The Alberta Court of Appeal in R v Navales, 2014 ABCA 70, was tasked with applying this law in Alberta.  At issue was how officers would use their training and experience, and a constellation of neutral “no win” behaviours on the part of the accused to form the objective grounds needed to find reasonable suspicion. The result has been what dissenting judges have referred to as a lowering of the standard to that of a generalized suspicion. Significantly, this line of decisions has been applied outside of the drug sniffing dog context, and even outside of the reasonable suspicion context, to other areas of criminal law in R v Canlas, 2014 ABCA 160, R v Ng, 2014 ABPC 62, and R v Tosczak, 2014 ABQB 86. Drug sniffing dog cases have proven to be an interesting area for developing criminal law jurisprudence regarding sections 8 and 9 of the Charter. Drug sniffing dog investigations by their very nature generally involve both a detention and a search. The Supreme Court found in R v Kang-Brown, 2008 SCC 18, that an officer would require a reasonable suspicion to perform the dog sniff search. Flash forward four years later, and both the Supreme Court, and the Alberta Court of Appeal are still trying to determine what will amount to a reasonable suspicion to perform the sniff search.  Reasonable suspicion requires a subjective belief that is objectively reasonable and subject to judicial scrutiny (Kang-Brown at paras 26, 75). The Supreme Court of Canada in Chehil found that an officer’s training and experience can provide the objective component necessary to form a reasonable suspicion (Chehil at paras 46, 47). Mackenzie dealt with “no win” behaviour, which is “characteristics that apply broadly to innocent people – he looked at me, he did not look at me” (Mackenzie, at para 71). Justice Moldaver found that this behaviour does have “some value” in forming reasonable suspicion when looked at as part of a constellation of factors assessed against the totality of the circumstances (at para 71).

There was also a strong dissent in Mackenzie written by Justice Lebel for himself, Chief Justice McLachlin, Fish J., and Cromwell J. The dissent highlighted a point from the Court’s previous decision of Kang-Brown, that after the fact judicial scrutiny must be rigorous as exclusion of evidence is the only check on police powers regarding breaches under sections 8, 9 or 10 of the Charter (Mackenzie, para 96). The dissent’s main concern was that the approach taken by the majority transformed the reasonable suspicion requirement into a requirement of only a generalized suspicion (at para 97). The dissent was concerned with courts and police draw

A Smart Decision – Access to Counsel for the Poor and Disabled in a Legal Aid Crisis

Thu, 09/04/2014 - 10:00am

By: Sarah Burton

PDF Version: A Smart Decision – Access to Counsel for the Poor and Disabled in a Legal Aid Crisis

Case commented on: R v Smart, 2014 ABPC 175

Access to justice advocates should to take a few moments to review R v Smart, 2014 ABPC 175, where the Honourable Assistant Chief Judge Anderson stayed proceedings against three accused persons who could not afford counsel, but did not qualify for Legal Aid. While such applications are not uncommon, the evidence considered in Smart extends far beyond the norm. This extensive evidence, coupled with Judge Anderson’s probing commentary on access to justice, places a welcomed spotlight on Alberta’s Legal Aid funding crisis. In Smart, Judge Anderson sought to provide concrete guidance to courts facing similar applications. While he accomplished this task, his engagement with access to justice issues may be the more lasting legacy of the judgment.

Facts

Three accused persons in unrelated cases were denied Legal Aid counsel because they exceeded the program’s financial limits. Each accused derived substantially all of their income from the Assured Income for the Severely Handicapped (“AISH”) program – $1,588.00/month (in one case $1608.00/month). They each submitted affidavits stating that they were denied legal aid, had little to no disposable income and could not afford to pay a lawyer. They all expressed discomfort self-advocating and communicating orally. With the exception of one accused, they had each made various unsuccessful efforts to retain counsel outside of Legal Aid.

Each accused was facing a charge of assault in addition to other offences (including possession of a weapon for a dangerous purpose, uttering threats, indecent exposure, breaking and entering, and breach of probation). Their defences were varied – self-defence, deceit by the alleged victim, and issues relating to mental capacity. Each accused was afflicted with at least one mental disorder (including Fetal Alcohol Spectrum Disorder, Attention Deficit Hyperactivity Disorder, and anxiety with a severe lack of memory). The Crown was seeking jail time in all cases.

The applicants sought to stay the proceedings against them pending the appointment of state counsel (commonly known as a “Rowbotham Application”). To increase efficiency, and to allow for extensive evidence on Alberta Legal Aid, the separate applications were consolidated.

The Nature of a Rowbotham Application

A Rowbotham Application (named for R v Rowbotham, [1988] OJ No 271, 1988 CanLII 147 (Ont CA)) is a conditional stay of proceedings granted where trial fairness is compromised by a lack of counsel. While the Charter does not contain a blanket right to legal representation, ss. 7 and 11(d) of the Charter demand that counsel be appointed if a case is so complex and serious that a fair trial cannot proceed without a lawyer. Judge Anderson aptly described the Rowbotham Order as a “pay or stay” system – either the government pays for state-funded counsel, or the charges against the accused are stayed (at para 17).

There are three basic elements of a successful Rowbotham Application: “(a) the accused must have sought the assistance of Legal Aid and been rejected, having exhausted any avenues of appeal, (b) the accused must be indigent, and (c) the charges must be sufficiently serious and complex to warrant judicial action” (at para 19). When jail time is sought, the case is often considered sufficiently serious (at para 149). Complexity requires an examination of the charge, the defences and the capacities of the accused (at para 161).

Evidence and Submissions

A typical Rowbotham Application focuses on the accused’s income and expenses, and an assessment of the case against him or her. In this regard, the accused persons in Smart submitted affidavit evidence of their financial circumstances, and the Court received summaries outlining the nature of the charges against them.

The affidavits were not ideal – they outlined each accused person’s estimated income and expenses with little (or no) exhibits to back up the figures. These estimates sometimes created the impression that the accused had a significant surplus of cash at the end of each month. Yet in oral examination, each accused indicated they never had extra money by month-end.

In addition to this typical evidence, Smart also considered the social restraints and background context of the applications. This evidence was provided through the CEO of Legal Aid (Ms. Suzanne Polkosnik Q.C.) and a member of the junior Edmonton criminal defence bar with experience working for Edmonton Student Legal Services (Ms. Tara Hayes). This additional evidence was candid and striking. It detailed the extent of Legal Aid’s limitations, and the relative lack of options for persons who (barely) do not qualify for Legal Aid. For present purposes, the following key points were discussed:

    • Legal Aid has three categories of service. As the level of service increases, the financial eligibility limit decreases. Only the third category (with the lowest eligibility cap) provides the full legal representation relevant to a Rowbotham Application (at para 26).
    • Financial eligibility guidelines are based on income and family size. Legal Aid officers have the discretion to exceed these guidelines by up to 15%, and no more. For a single person seeking full legal representation, the financial guideline is $1,348/month, and $16,170/year (both thresholds must be satisfied). Persons receiving AISH earn $38.00 more than the 15% discretionary threshold permits (at paras 3, 30).
    • $2.5 million of Legal Aid’s budget is a “contingency fund” to cover unpredicted costs, which includes court-ordered legal aid representation. This amount is in peril of being exhausted, at which point money will come from a different source of Legal Aid’s budget (at paras 23, 24).
    • Student Legal Services is a student legal assistance office. The students operate as agents, not lawyers for an accused. They only act on summary conviction offences, and they do not act in cases where there is a foreseeable risk of jail (at para 43).
    • For a non-complicated matter, Ms. Hayes seeks a retainer of $1,500. Her quoted fee for an uncomplicated one-day trial was approximately $3,000 (at para 44).

The Decision

Judge Anderson granted the three Rowbotham Applications. He did this despite significant shortages in the affidavit evidence, unexplored appeals to Legal Aid, and questions regarding the prospect of jail time.

With regards to the affidavit evidence, the applicants’ quoted figures were often lacking in both accuracy and documentary corroboration. In some cases, they failed to demonstrate that the accused could not afford counsel. Judge Anderson held that the affidavits were made with best efforts, but that they (significantly in some cases) understated the applicants’ expenses (see e.g. at paras 194, 195). Each applicant lived below the poverty line and testified that they had no money at the end of the month. The Court accepted this evidence. Further, the Crown’s complaints did not resonate, because it failed to question these inaccuracies while cross examining the applicants (see e.g. at para 184).Thus, these clearly lacking affidavits (somewhat counterintuitively) lent credence to the applicants’ claims that they required counsel.

Judge Anderson was not persuaded that the charges were insufficiently serious to require counsel. The Crown argued in one case that jail time, while sought, was unlikely to be ordered. Judge Anderson was not swayed, noting that the Crown’s decision to seek jail time caused the accused’s Student Legal Services counsel to withdraw (at para 199). The fact that jail time was possible, though unlikely, seemed to further support the applicant’s position that he needed counsel.

Judge Anderson’s decision was also impacted by the accused persons’ clear (though yet unproven) brain injuries and/or mental disorders. While there was no proof that the persons suffered from the mental disorders, Judge Anderson refused to ignore “the elephant in the room” (at para 206). A person under a mental disorder is less able to defend themselves adequately. Moreover, proving that a mental disorder impacted the accused persons’ mens rea (as in the FASD case) is very difficult and complex. Being under a mental disorder added weight to the argument that counsel was needed.

Lastly, Judge Anderson was not swayed by the fact that unexplored (though likely futile) efforts to retain counsel existed. Each of the accused persons in some respects failed to (a) follow through the Legal Aid appeal process, (b) contact community and support groups, and/or (c) exhaust the gambit of private criminal defence lawyers. Judge Anderson drew from Ms. Polkosnik’s and Ms. Hayes’ evidence to conclude that these efforts had no realistic chance of success. The accused persons had no money to retain counsel and an appeal to Legal Aid would plainly not succeed. In the interests of the timely administration of justice, the accused persons should not have to jump through useless hoops to be appointed counsel (at para 203).

Commentary on Access to Justice

While the Smart decision will undoubtedly be a valuable tool when adjudicating future Rowbotham Applications, its utility will extend beyond those borders. Anyone grappling with access to justice (accused persons, civil litigants and academic commentators) will find value reviewing this decision, which touches on several recurring themes in access to justice commentary. A few of these themes are explored below.

The Focus on Practical Reality 

Judge Anderson repeatedly emphasized the need to focus on the practical reality that courts and accused persons face when attempting to access justice. The Legal Aid budget has created a crisis for the poor and middle class to access counsel. While lawyers are comfortable making legalistic arguments grounded in evidentiary thresholds, precedents and technical language, this approach can frustrate rather than facilitate justice. As such, when faced with access to justice concerns, judges should be focused on the realistic lived experience of the applicants. For example, Judge Anderson held that:

    • The technical possibility that an accused could, over a series of months, possibly save enough to retain a lawyer should not lead to a rejection of counsel. Judges must remain focused on the accused’s ability to retain counsel in a realistic timeframe (at paras 7, 197, 205).
    • Applying precedents from complex cases with sizable assets may only muddy the water. The vast majority of Rowbotham Applications deal with those who are “poor…marginalized…at a low point in life” and “tend to be those just getting by” (at para 130). The precedents from more complicated cases may not be of use for the majority of these applications (para 96, referring to R v Malik, 2003 BCSC 1439).
    • Just because an affidavit would, on its face, suggest that counsel could be retained, a judge should look at the substance of the material and inquire if those figures are realistic. In doing so, judges should not ignore the presence of mental injuries or disorders (at para 194).
    • Judges should not demand that accused persons jump through useless hoops before their application is granted. Legal Aid’s discretion is set in stone. Until funding changes, there is no point delaying justice while the accused pursues a doomed appeal (at paras 176, 203, 204).
    • When assessing case complexity, judges often underestimate or ignore the complexity of a defence. Cases that first appear simple can become exceptionally complex when a defence is raised. In these cases, judges will find themselves drawing heavily on basic life and judicial experience (at paras 164, 167).

Underlying this focus on practicality is the recognition that Rowbotham applicants are often self-represented. They are often the least equipped to be making their application, and they often arise in the least favourable setting (a faced-paced, high volume docket court) (at para 115).

In such circumstances, judges must recognize that self-represented parties should not be treated as though they are experienced counsel. Assistance is needed, but at the same time, judges have to respect the boundaries of the adversarial system (at para 113).

The Proper Role of Judges

Stemming from this, some have suggested that the risk of trial unfairness without counsel is overblown because judges can ensure that a trial proceeds fairly. Judge Anderson cautioned against this approach for two reasons. First, it ignores the role of defence counsel in creating a fair playing field. Judges cannot advise an accused on strategy or preparation – they can only address the unfairness present in the courtroom. The trial is only the tip of the iceberg when launching a criminal defence (at paras 111-112).

Second, relying on the trial judge to protect self-represented parties draws the judge “into the fray” and risks creating a reasonable apprehension of bias (at para 110). Judges have to remain impartial and neutral in the courtroom. They must not only ensure a fair trial proceeds, but that the trial appears fair in the eyes of the public. When judges get pushed out of the neutral role, the appearance of fairness to in the courtroom can be compromised.

The Costs Associated with Legal Aid

Discussions about Legal Aid must proceed unburdened by false assumptions. In particular, people generally believe that Legal Aid and Rowbotham Orders are a drain on taxpayer dollars. This belief is built on two false notions.

First, Rowbotham Orders are not an additional cost on taxpayers – the money to pay a lawyer comes out of Legal Aid’s existing budget (at para 14). Second, the assumption that Legal Aid costs money fails to consider the costs incurred by the justice system when counsel is not present. As Judge Anderson noted:

[15] Anyone on the front lines of the justice system will recognize the important role that defence counsel plays in allowing matters to proceed efficiently. Judicial experience shows that when accused persons are not represented, the number of appearances tends to increase, the number of adjourned trials tends to increase and the length of trials tends to increase, each of which involves significant cost to the government through judicial, prosecutorial, clerical and security costs.

The Responsibility to Support Indigent Accused Persons  

The Crown attempted to argue that Rowbotham Applicants must seek out pro bono counsel or financial assistance from friends or family before being appointed counsel. Judge Anderson unequivocally rejected both suggestions.

As for pro bono counsel, Judge Anderson found it insulting to suggest that the criminal bar ought to be burdened with fulfilling the government’s obligation to provide counsel. He stated:

[140]  The Court categorically rejects this expectation as a requirement to show need in a Rowbotham application. As is well known to the courts, no segment of society does more to assist the indigent facing prosecution by the state, on a pro bono basis, than the defence bar.

[141]  Providing access to justice is the obligation of the government, not good-willed citizens. The Charter protects against state failures not the failure of citizens to make up for state short-comings. With the greatest of respect to the contrary views expressed in some of the jurisprudence, this expectation and its underlying assumptions are insulting to both the impecunious and to the Bar.

He was equally affronted by the suggestion that an accused must seek money from family and friends:

144 … [I]t would be dangerous and in many cases unfair to both the accused and the third parties, if the Court was to make [requesting money from family or friends] a pre-requisite to a Rowbotham Order.

[145]      It is the state that is seeking to hold the accused to account for his or her actions and ensure that the prosecution of that action is fair. Prosecutions are not brought by individuals and it would not be fair to require individuals to fund a matter that is not being brought on their behalf but rather on behalf of society at large.

These statements firmly place the responsibility for ensuring counsel on the government. While members of society can step in to assist a particular person in need, this cannot minimize the government’s Charter obligation to guarantee a fair trial.

Concluding Thoughts: Rowbotham Orders and the Future of Civil Legal Aid

Given the position of the accused persons in Smart, I would predict that most readers approve of Judge Anderson’s order to appoint them counsel. This is a sentiment I share, but one which is complicated by my concerns for the civil legal aid system.

The vast majority of Rowbotham Orders are made in criminal cases. Persons accused in the criminal legal system do not enjoy a blanket constitutional right to counsel, but Rowbotham Orders provide a solid footing to prevent trial unfairness. This is not so in the civil system. Section 11(d) of the Charter does not apply outside the criminal sphere. As jail time is not on the table in civil actions, they are often not serious enough to trigger s. 7 Charter protections. As such, civil litigants largely operate without any right to counsel (for an exception to this rule, see New Brunswick (Minister of Health and Community Services) v G. (J.), [1999] 3 SCR 46).

Presently, Legal Aid offers assistance to parties outside the criminal system under extremely restrictive financial guidelines. This would include (for example) family law, immigration, foreclosure, bankruptcy, and general civil actions. However, as Ms. Polkosnik’s testimony made clear, the increasing number of Rowbotham Orders have exhausted Legal Aid’s contingency fund. Future orders must draw from another source of Legal Aid’s budget. While not stated, one may assume that the money to fund Rowbotham Orders is coming at the expense of the civil Legal Aid budget.

Any civil litigator can attest to the fact that the litigation system is flooded with self-represented parties who plainly cannot afford a lawyer. This slows the justice system while these litigants attempt to navigate a complex and seemingly impenetrable maze of legal procedures. For the self-represented litigant, facing (for example) financial ruin, a child custody battle, or losing their home, “access to justice” is an empty phrase that means nothing for those without money or connections (The Canadian Bar Association, Access to Justice Committee, Reaching Equal Justice: An Invitation to Envision and Act, (Ottawa: The Canadian Bar Association, November 2013) at Part 1).

So, while Judge Anderson’s decision may be applauded, we cannot ignore the fact that criminal Rowbotham Orders are merely re-shuffling the burden of an underfunded Legal Aid system to other unfortunate people with fewer protections.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

Deference, Discretion, and Adjudicative Substitution

Wed, 09/03/2014 - 10:00am

By: Camille Sehn

PDF Version: Deference, Discretion, and Adjudicative Substitution

Case commented on: E.G. v Alberta (Child, Youth and Family Enhancement Act, Director), 2014 ABCA 237

Editor’s Note: The University of Calgary Faculty of Law runs a fantastic legal clinic, Student Legal Assistance (SLA), that employs several students during the summer. This year’s crop of students was encouraged by SLA’s Executive Director, Michelle Christopher, to submit posts to ABlawg in the criminal and family law areas. This is the second post in the series; the first is available here and there will be more to come.

**********************************************************************

In a recent decision from the Court of Appeal of Alberta, the Honorable Mr. Justice Jack Watson engages in a discussion around judicial discretion in an often nuanced and complex area of the law. At trial, the Director of Child and Family Services (“the Director”) applied for a Permanent Guardianship Order (“PGO”) of two twin boys; both parents contested the PGO. After 16 days of evidence at trial in Provincial Court, the Honorable Judge L.T.L. Cook-Stanhope concluded that there was gross abuse of the twins based on evidence of a pattern of maltreatment and poor attitudes towards parenting. Judge Cook-Stanhope found that the parents presented a substantial risk of further physical and emotional harm, and relied on the testimony of several witnesses to support this apprehension of substantial risk if the twins were to be returned to their parents. She granted the PGO on this basis. (E.G. v Alberta (Child, Youth and Family Enhancement Act, Director), 2013 ABPC 311 at para 224)

On appeal to the Court of Queen’s Bench, the Provincial Court decision was reversed on the grounds that the Court acted on a wrong principle and disregarded material evidence (E.G. v Alberta (Child, Youth and Family Enhancement Act, Director), 2014 ABQB 406 at para 3). The Honorable Mr. Justice G.C. Hawco disagreed with the weight Judge Cook-Stanhope placed on expert testimony at trial, and found that she disregarded expert written reports which suggested that the parents had motivation and capacity for positive change. According to section 34(1)(C) of the Child, Youth and Family Enhancement Act, RSA 2000, c. C-12 (“the Act”), a PGO is only to be used in most extreme cases where there is no reasonable hope of returning the child to the parents in a reasonable time (ABQB at para 11). As such, according to Justice Hawco, these reports should have been considered by Judge Cook-Stanhope because they appear to provide some support that a PGO may not be necessary. Justice Hawco also disagreed with the definition of abuse adopted by Judge Cook-Stanhope because there was some inconsistency in the older children’s reports of the abuse, and because the twins did not display symptoms of trauma (ABQB at para 14). Essentially, Justice Hawco re-evaluated the evidence, the weight of the written reports, and dispensed with Judge Cook-Stanhope’s finding that the twins were victims of abuse.

The Director applied to the Court of Appeal for a stay of Justice Hawco’s decision pending appeal, which was granted by Justice Watson. In his reasons, Justice Watson was careful to focus on the grounds of the appeal rather than allow his own perception of the evidence to guide his decision. He also made certain to avoid any direction on custody and to recuse himself from participating in the appeal. This is important given the issues being argued: how much deference should be given to the fact finding process at trial? And when does an appeal decision arise from adjudicative substitution rather than a palpable and overriding error?

The written reports were deemed admissible at trial, but they were not relied upon by Judge Cook-Stanhope; she made it quite evident that she relied heavily on the testimony of the witnesses who contradicted the written reports (ABCA at para 4). Justice Watson discussed the problem with one report in particular, by Dr. Vellet. On appeal, Justice Hawco appeared to have relied almost completely on the opinion expressed in this report by reversing Judge Cook-Stanhope’s decision, yet Dr. Vellet was not called by the parents and, as such, was not open to cross examination. This is troubling because the authors of written reports are not then available to answer questions about the parents’ past actions or speak to other pertinent issues around risk (ABCA at para 15).

The Court of Appeal stayed the Queen’s Bench decision because Justice Hawco appeared to have re-weighed the evidence from trial. Justice Hawco opined about the prejudicial influence of evidence presented at trial about the treatment of the older, adopted children, and claimed that Judge Cook-Stanhope failed to consider whether the adopted children could have been treated differently than the biological twins. Further, Justice Hawco took issue with the definition of abuse adopted by Judge Cook-Stanhope, despite the evidence given at trial by an expert in Infant Mental Health who suggested that mere exposure to the abuse of others has the same effect on a child’s development and often results in the same problematic outcomes (ABPC at para 122).  This re-weighing of the expert reports indicated that while Judge Cook-Stanhope did not find the written reports reliable, Justice Hawco found them to be “material evidence” which had been disregarded, resulting in grounds for appellate review according to M.T. v Alberta (Director of Child Welfare), 2005 ABCA 125 at para 241.

While a PGO is the worst outcome a parent can receive under the Act, the decision making process for custody and guardianship must always be guided, firstly, by the best interests of the child or children in question. This process relies upon evidence presented by all parties to an action, and according to section 34(1) of the Act, the evidence must show there is no reasonable possibility of returning the children to their parents if a PGO is to be granted. By staying the decision of Justice Hawco, it seems that the Court of Appeal has indicated that if there is expert evidence that a party wishes to rely upon in order to cast some doubt on the necessity of a PGO, that evidence must be made available for cross examination.

The three part test adopted in Alberta regarding a stay is arguableness, irreparable harm and balance of convenience, though the test has been modified in cases involving children to include consideration of their best interests (B(C) v C(P), 2003 ABCA 321). Arguments for the parents against the stay focused on Judge Cook-Stanhope’s fact finding process and exercise of discretion under section 34(1) of the Act, but Justice Watson found the Director’s arguments on procedural and evidentiary errors made by Justice Hawco alongside preserving the physical and psychological stability of the children much stronger (I refer to section 34(1) because although Justice Watson cites section 36(1) of the Act (at para 27), section 36(1) was repealed in 2003, and so I assume this to be a typographical error). Justice Watson found the factors all favored arguments by the Director for the stay, because it would preserve the integrity of the fact finding process and prevent the yo-yo effect where children go back and forth between their parents and the state (ABCA at para 28).

Ultimately, expert evidence submitted without an opportunity for cross examination should carry less weight at trial. The complexities of childhood development and parental patterns of abuse do not exist in a vacuum. Therefore, expert opinions cannot be relied upon heavily without a full and complete opportunity to explore those opinions within the entire context of the issue at trial. Even when that opportunity is used, according to this recent development, the trial judge must be granted deference in analyzing and weighing the expert evidence alongside all other evidence to make a decision. In reversing Judge Cook-Stanhope’s decision, it appears that Justice Hawco erroneously considered the written reports “material evidence”, and further, substituted his own fact finding process on appeal. Justice Watson rightfully stayed this decision pending appeal, but the decision also highlights the potential issues on appeal in matters under the Act wherein an appellate court should support and uphold the integrity of the discretionary process of fact finding at trial, and avoid the occurrence of adjudicative substitution.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

 

New Parameters for Family Status Adopted in Alberta

Tue, 09/02/2014 - 10:00am

By: Linda McKay-Panos 

Case Commented on: Clark v Bow Valley College, 2014 AHRC 4

Recently, human rights decisions in federal tribunals and courts have adopted a broader definition of “family status” as meaning more than just one’s relationship to another person, and recognizing childcare responsibilities. Rights groups have been positive about this development, but perhaps some employers are concerned. The leading case, Canada v Johnstone, 2014 FCA 111, was discussed in previous posts (see here). Alberta’s Human Rights Tribunal has now adopted and applied this jurisprudence in Alberta.

Clark was a nursing instructor with Bow Valley College. She requested and was granted maternity leave from February 1, 2010 to January 31, 2011. She went on sick leave in November 2009, and her child was born nearly seven weeks premature (on January 2, 2010 instead of February 21, 2010 as expected). After the premature birth of the child, there was no communication about the start or end date of the maternity leave. One letter from the College about benefits referred to Clark’s maternity leave as being from February 1, 2010 to January 31, 2011 (as was first planned).

Clark and her family lived in Linden, AB, which is approximately 100 km from Calgary. In mid-November, 2010, she became aware that she had been placed on the instructor schedule beginning January 3, 2011. She contacted Bow Valley College to advise that she had childcare in place beginning February 1, 2011, and was not having any success in obtaining childcare before then. Clark offered to take some of her vacation leave in January. Bow Valley College denied her any leave past January 10, 2011, and when she did not report for work on January 13, 2011, they deemed Clark to have resigned from her position and she was terminated (paras 4, 5).

Clark argued that Bow Valley College was aware that she had no options for childcare, and that the denial of leave was discrimination on the basis of family status.

Bow Valley College argued that under the collective agreement, Clark was allowed only 52 weeks maternity leave. They also said that they did not discriminate against Clark; rather, there was a shortage of nursing faculty and they could not locate coverage. Thus, the denial of a vacation request was based on Bow Valley College’s operational requirements and not Clark’s family status. Further, Bow Valley College had attempted to accommodate Clark by providing her with two alternative childcare options. Thus, Bow Valley College submitted that it did not discriminate against Clark, and further, they had accommodated her to the point of undue hardship (para 6).

Bow Valley College also argued that Clark had not provided sufficient evidence about her husband’s income generating activities or his ability to provide childcare (para 7).

Clark testified that she was the primary breadwinner and caregiver for her family. Her husband contributed about $100 per month to the family income. She did not consider her husband to be a suitable caregiver because of his health issues (para 23). Further, there were no viable extended family caregivers (paras 15, 16). In addition, the two alternatives for childcare proposed by Bow Valley College were not feasible: one did not accept children young enough and the other one involved communal care which would compromise the child’s immune system (para 18). Clark did not have the money to purchase a car seat suitable for the commute to Calgary in her vehicle. Both daycare organizations were not open early enough to accommodate her work schedule (para 19).

The Chair of the Tribunal, Sharon Lindgren-Hewlett, noted that correspondence from Bow Valley College to Clark did not address her childcare concerns. They did not request a meeting to work out the problem or understand Clark’s concerns (para 30).

The Chair noted also that Bow Valley College had a “fairly high standard of care” to confirm Clark’s leave dates (para 38).  Thus, there was a mutual mistake on the part of both parties with respect to the leave dates.

In determining whether there was prima facie discrimination, the Chair relied on the factors set out by the Supreme Court of Canada in Moore v British Columbia (Education), 2012 SCC 61. The Chair indicated that (para 42):

A complainant is required to show three elements: they have a characteristic protected from discrimination; they have experienced an adverse impact with respect to employment; and the protected characteristic was a factor in the adverse impact.

Bow Valley College and Clark differed on whether a prima facie case of discrimination on the basis of family status was made out. During the course of the Clark hearing, the Federal Court of Appeal released Johnstone, supra, and Canadian National Railway Company v Seeley, 2014 FCA 111. Both sides were requested to make submissions based on the four-part test set down in these cases for prima facie discrimination in family status childcare obligation cases. The four factors are set out in the following excerpt from Johnstone:

[93] I conclude from this analysis that in order to make out a prima facie case where workplace discrimination on the prohibited ground of family status resulting from childcare obligations is alleged, the individual advancing the claim must show (i) that a child is under his or her care and supervision; (ii) that the childcare obligation at issue engages the individual’s legal responsibility for that child, as opposed to a personal choice; (iii) that he or she has made reasonable efforts to meet those childcare obligations through reasonable alternative solutions, and that no such alternative solution is reasonably accessible, and (iv) that the impugned workplace rule interferes in a manner that is more than trivial or insubstantial with the fulfillment of the childcare obligation.

[94] The first factor requires the claimant to demonstrate that a child is actually under his or her care and supervision. This requires the individual claiming prima facie discrimination to show that he or she stands in such a relationship to the child at issue and that his or her failure to meet the child’s needs will engage the individual’s legal responsibility. In the case of parents, this will normally flow from their status as parents. In the case of de facto caregivers, there will be an obligation to show that, at the relevant time, their relationship with the child is such that they have assumed the legal obligations which a parent would have found.

[95] The second factor requires demonstrating an obligation which engages the individual’s legal responsibility for the child. This notably requires the complainant to show that the child has not reached an age where he or she can reasonably be expected to care for himself or herself during the parent’s work hours. It also requires demonstrating that the childcare need at issue is one that flows from a legal obligation, as opposed to resulting from personal choices.

[96] The third factor requires the complainant to demonstrate that reasonable efforts have been expended to meet those childcare obligations through reasonable alternative solutions, and that no such alternative solution is reasonably accessible. A complainant will, therefore, be called upon to show that neither they nor their spouse can meet their enforceable childcare obligations while continuing to work, and that an available childcare service or an alternative arrangement is not reasonably accessible to them so as to meet their work needs. In essence, the complainant must demonstrate that he or she is facing a bona fide childcare problem. This is highly fact specific, and each case will be reviewed on an individual basis in regard to all of the circumstances.

[97] The fourth and final factor is that the impugned workplace rule interferes in a manner that is more than trivial or insubstantial with the fulfillment of the childcare obligation. The underlying context of each case in which the childcare needs conflict with the work schedule must be examined so as to ascertain whether the interference is more than trivial or insubstantial.

The Chair concluded that all four elements of the Johnstone test for childcare obligations under family status fall within the first element of Moore test for prima facie discrimination. The Chair concluded that Clark clearly satisfied the first two Johnstone factors, and the fourth factor as well (not a trivial interference), but that Clark must also demonstrate a bona fide childcare problem remains and that she had not been able to locate a solution on her own (the third factor) (para 49).

Unlike Johnstone or Seeley, supra, this case was not about a permanent change in hours, or a permanent shift adjustment, or a request for further benefits, or increased payment. The matter at hand merely involved a mistake in the return to work date, followed by a denial of a three-week vacation, followed by termination of employment. The Chair determined that the issue to be determined was whether Clark expended reasonable efforts to locate reasonable childcare solutions, including her partner’s availability and suitability to care for the child (para 50). The Chair also concluded that Clark was actually given between six and 21 days’ notice (over a holiday season) to find childcare or be considered to have abandoned her job.

Although Bow Valley College asserted that Clark’s credibility was an issue because she did not call her partner as a witness, her own testimony established that it was not reasonable for her partner to quit his work to care for the child for one month.  Thus, Clark was truthful and reasonable in not asking her partner to quit his employment because of a three-week childcare challenge (para 56). Further, Clark faced many obstacles in obtaining childcare because of the mistaken return to work date.

In applying the three factors from Moore, supra, the Chair concluded that Clark had a bona fide childcare problem which satisfied the protected ground of family status; the neutral rule of returning to work on January 3, then January 10, adversely impacted Clark; and, the only reason for her absence from work was the childcare problem, thus demonstrating the connection of the adverse impact to the ground of family status (para 65). Thus, prima facie discrimination on the basis of family status was established.

Bow Valley College defended its position, relying on the Alberta Human Rights Act, RSA c A-25.5, section 11, reasonable and justifiable discrimination, and section 7(3) bona fide occupational requirement. Bow Valley College argued that there was a shortage of nursing faculty at the time and that they believed that Clark was abandoning her position because she believed she had no childcare options (paras 66, 67).

The Chair noted that when Clark communicated her childcare problem, she was, for all practical purposes, requesting an accommodation. In this situation, Bow Valley College did not seek any information upon which they could conduct an assessment; nor was there any consideration of the information provided by Clark at the time. No collaboration or alternative approach was explored with Clark. Clark had 35 days of accrued vacation, and granting her vacation leave would have been a possible accommodation (paras 70-75).

Although Bow Valley College submitted that its operational requirements did not permit any additional absence, there was no demonstrated undue hardship for Bow Valley College to have implemented a shared instructor situation as they did in other cases. Further, Bow Valley located an instructor to replace Clark without even advertising for one (paras 77, 78). Thus, Bow Valley College failed to accommodate Clark to the point of undue hardship, and the defences failed.

The remedy awarded to Clark included $15,000 for injury to dignity, and lost wages for the period of February to May 1, 2011 (to be calculated by the parties).

The overall implication of these cases (including Clark) is that courts and tribunals are clearly recognizing the continued gendered aspects of childcare arrangements. Women, who are increasingly in the workplace, also shoulder the bulk of responsibility for childcare. To discriminate against a worker because she has childcare challenges will amount to discrimination on the basis of family status. Parents’ childcare concerns will need to be accommodated to the point of undue hardship.

Access to Justice: The DIY Index

Thu, 08/28/2014 - 10:00am

By: John-Paul Boyd

Editor’s Note: John-Paul Boyd, the Executive Director of the U of C-affiliated Canadian Research Institute for Law and the Family (CRILF), started a new blog in August on Access to Justice in Canada. John-Paul will be cross-posting on ABlawg from time to time and blogging on family law decisions (see also his blog JP Boyd on Family Law). This first post is an index to five separate entries on DIY access to justice approaches originally posted on Access to Justice in Canada.

*******************************************************************************************

DIY #1: Write and share plain-language information on the law

Prepare and distribute handouts with clear information about the law and dispute resolution processes. Handouts could cover topics including the substantive law on different issues, management of common litigation tasks and tips for successful mediation. Leave them in a brochure rack in your reception area; give copies to colleagues and to pro bono and community legal clinics; give them to social service providers such as abused women’s centres and immigrant settlement agencies.

DIY #2: Work with others and other’s work

Connect with a few of the social service agencies in your neighbourhood, find out where the holes are in their library of legal resources, and fill them. Think and write about the law in a way that addresses the unique legal needs and realities of each group’s target population. Work with community media and larger social service groups; these generally have a broader reach and better funding, and the work you do often goes much further.

DIY #3: Talk to your community

Get in touch with the libraries, community centres and social service groups in your area and arrange to provide one or more public lectures; public talks are a rewarding, enriching and engaging way of improving access to justice. The range of topics you can address is unlimited and could include introductions to court processes, alternatives to court, landlord tenant law, wills and estates, the basics of family law, and anything else that could of interest to the people you are talking to. Providing handouts gives the community group and the people at your talk an additional resource.

DIY #4: Unbundle your services, reinvent your billing model

Working on an unbundled basis is a great way to maintain a remunerative practice while offering legal services that are more accessible than services offered on a comprehensive, billable-hour basis, however few lawyers offer such services. Unbundling gives clients the services they select on fixed or predictable prince and within a defined time period; it gives lawyers a less stressful practice with a lower likelihood of mounting accounts receivable.

DIY #5: Do more work on a flat rate basis

Offering services on a flat rate basis is another way to maintain a profitable practice while improving access to justice. Under this model, the client can pick and choose which and how much of a lawyer’s services he or she will buy, at a fixed rate which is determined up front. The client and the lawyer are protected from the client’s frustration if a legal issue is not resolved before his or her resources are exhausted. The lawyer gets a file with a fixed scope of required labour and a minimal potential of becoming a dog file, payment up front and a minimal likelihood of collections issues, and a file free from the tyranny of recording time.

The Relevance of Sattva for Appeals from Arbitration Awards in Alberta

Tue, 08/19/2014 - 10:00am

By: Jonnette Watson Hamilton

PDF version: The Relevance of Sattva for Appeals from Arbitration Awards in Alberta

Case commented on: Sattva Capital Corp v Creston Moly Corp, 2014 SCC 53 (CanLII)

The Supreme Court’s decision in Sattva Capital Corp v Creston Moly Corp has quite rightly received a great deal of attention. It has attracted notice in contract law circles for changing the law by holding that contractual interpretation involves questions of mixed fact and law, and not questions of law (see e.g. “Contract interpretation is no longer a question of law”, “A blockbuster decision in contractual interpretation” and “SCC issues ‘big change’ to contract law – Sattva gives last word to trial judges, arbitrators”). And, because the precedent-setting decision arose from an arbitration hearing in British Columbia, it has also attracted commentary more focused on the arbitral aspects (see e.g. “Finally, the Supreme Court of Canada puts some finality into Arbitrations” and “Supreme Court of Canada Limits the Right to Appeal Commercial Arbitral Decisions on Issues of Contractual Interpretation”). Because the British Columbia arbitration legislation that facilitated and regulated the arbitration in Sattva is unlike that in the rest of common law Canada, I will focus on the arbitration aspects of the decision and then explore the difference the Sattva decision may make in arbitrations in Alberta (and in Ontario, Saskatchewan, New Brunswick, Prince Edward Island, Manitoba and Nova Scotia, all of which also adopted the Uniform Law Conference of Canada’s Uniform Arbitration Act (1990)).

The Supreme Court of Canada decision

The initial dispute was about a US$1.5 million finder’s fee that Creston Moly was contractually obligated to pay Sattva Capital. The finder’s fee was payable in Creston shares. The parties disagreed about which date their contract required them to use to determine the price of the Creston shares. Depending upon the date chosen, Sattva would either get 11,460,000 shares or 2,454,000 shares. The arbitrator agreed with Sattva’s interpretation of the parties’ contract and awarded Sattva 11,460,000 Creston shares. Creston wanted to appeal.

Under British Columbia’s Arbitration Act, RSBC 1996, c 55, s 31, appeals of arbitration awards are only available on questions of law and only with either the consent of all the parties to the arbitration or with leave of the court:

 31 (1) A party to an arbitration … may appeal to the court on any question of law arising out of the award if

(a) all of the parties to the arbitration consent, or

(b) the court grants leave to appeal.

(2) In an application for leave under subsection (1) (b), the court may grant leave if it determines that

(a) the importance of the result of the arbitration to the parties justifies the intervention of the court and the determination of the point of law may prevent a miscarriage of justice,

(b) the point of law is of importance to some class or body of persons of which the applicant is a member, or

(c) the point of law is of general or public importance.

The British Columbia Supreme Court denied leave to appeal, holding there was no question of law. However, the British Columbia Court of Appeal held that leave should be granted and sent the dispute back to the Supreme Court for a decision on the merits of the appeal. The Supreme Court held the arbitrator’s decision was correct and dismissed the appeal. However, the Court of Appeal held the arbitrator’s decision was incorrect and allowed the appeal. The Supreme Court of Canada granted leave to appeal from both of the Court of Appeal’s decisions and, in the end, agreed with the arbitrator and the British Columbia Supreme Court.

The Supreme Court acknowledged that “[h]istorically, determining the legal rights and obligations of the parties under a written contract was considered a question of law” (para 43). However, the unanimous decision determined that this historical approach should be abandoned (para 50). Henceforth, “[c]ontractual interpretation involves issues of mixed fact and law as it is an exercise in which the principles of contractual interpretation are applied to the words of the written contract, considered in light of the factual matrix” (para 50). The Court acknowledged that it might be possible to extricate a question of law from the mixed fact and law context of contractual interpretation — for example, the application of an incorrect principle, the failure to consider a required element of a legal test, or the failure to consider a relevant factor (para 53) — those instances “will be rare” (para 55).

In this case, no question of law was raised. As a result, no appeal was available from the arbitrator’s award.

While that was enough to dispose of the appeal, the Supreme Court went on to consider one of the three factors in section 31(2) that have to exist before leave can be granted and whether leave to appeal should have been granted in this case, assuming the issue had been a question of law. First they looked at whether an appeal would have been necessary to prevent a miscarriage of justice, as specified by section 31(2)(a). The Court held that in order to meet that standard, “an alleged legal error must pertain to a material issue in the dispute which, if decided differently, would affect the result of the case” (para 70). In other words, the appeal had to have some possibility of succeeding (para 71); without considering the full merits of the case (para 72), it had to be clear that an argument that the arbitrator’s decision was unreasonable had merit (para 74).

Second, the Court looked at what considerations should factor into a court’s exercise of the residual discretion in section 31(2). British Columbia courts had determined that the words “may grant leave” in section 31(2) meant a court can refuse leave to appeal even when one of the three requirements listed in section 31(2) has been satisfied. The Supreme Court agreed a residual discretion existed (without discussing the point), but disagreed with the list of factors to be considered in exercising that discretion as set out in the leading case of British Columbia Institute of Technology (Student Assn.) v. British Columbia Institute of Technology, 2000 BCCA 496 (CanLII), 2000 BCCA 496. The list was overly broad, extending beyond the historical bases for refusing discretionary relief — the parties’ conduct, the existence of alternative remedies, and any undue delay (para 87) — and repeating factors already considered in the section 31(2) analysis.

The Court’s final comment on appeals from arbitration awards was about the standard of review.  Here the Court noted:

Appellate review of commercial arbitration awards takes place under a tightly defined regime specifically tailored to the objectives of commercial arbitrations and is different from judicial review of a decision of a statutory tribunal. For example, for the most part, parties engage in arbitration by mutual choice, not by way of a statutory process. Additionally, unlike statutory tribunals, the parties to the arbitration select the number and identity of the arbitrators. These differences mean that the judicial review framework developed in Dunsmuir v. New Brunswick, 2008 SCC 9 (CanLII), 2008 SCC 9, [2008] 1 S.C.R. 190, and the cases that followed it is not entirely applicable to the commercial arbitration context (para 104, emphasis added).

It is refreshing to see the Court acknowledge that appellate review of commercial arbitration awards is different from judicial review of decisions of statutory tribunals. This point is too often forgotten; see my post “Arbitration is not Administrative Law.” However, in subsequent paragraphs, the acknowledged differences do not seem to play any role. For example, all of the precedents cited on the standard of review point are administrative law precedents: Dunsmuir v. New Brunswick, 2008 SCC 9 (CanLII), 2008 SCC 9, [2008] 1 SCR 190; Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association, 2011 SCC 61 (CanLII), 2011 SCC 61, [2011] 3 SCR 654, and Newfoundland and Labrador Nurses’ Union v. Newfoundland and Labrador (Treasury Board), 2011 SCC 62 (CanLII), 2011 SCC 62, [2011] 3 SCR 708, citing David Dyzenhaus, “The Politics of Deference: Judicial Review and Democracy”, in M. Taggart, ed., The Province of Administrative Law (1997) 279.

As for the standard of review itself, the Court held “[i]n the context of commercial arbitration, where appeals are restricted to questions of law, the standard of review will be reasonableness unless the question is one that would attract the correctness standard, such as constitutional questions or questions of law of central importance to the legal system as a whole and outside the adjudicator’s expertise” (para 106, emphasis added). And, in analyzing whether an arbitrator’s decision was reasonable, “it is permissible for reviewing courts to supplement the reasons of the original decision-maker” (para 110).  Indeed, the Court relies almost entirely on the reasons offered by Justice Armstrong of the British Columbia Supreme Court (paras 107-119). This is arbitration’s version of the administrative law point about deference to “the reasons offered or which could be offered in support of a decision” (Dunsmuir at para 48), i.e., the result is correct because reasons could have been offered in support.

Application to Alberta

In Alberta, the equivalent of British Columbia’s section 31 is section 44 of the Arbitration Act, RSA 2000, c A-43:

44(1) If the arbitration agreement so provides, a party may appeal an award to the court on a question of law, on a question of fact or on a question of mixed law and fact.

(2)  If the arbitration agreement does not provide that the parties may appeal an award to the court on a question of law, a party may appeal an award to the court on a question of law with leave, which the court shall grant only if it is satisfied that

(a) the importance to the parties of the matters at stake in the arbitration justifies an appeal, and

(b) determination of the question of law at issue will significantly affect the rights of the parties.

Both the British Columbia and the Alberta legislation make appeals of arbitration awards on questions of law available if the parties consent or if the court grants leave to appeal. The Alberta statute allows parties more latitude to fashion their own agreement about appeal rights, not restricting them to questions of law.

Both the British Columbia and the Alberta legislation limit a court’s discretion to grant leave by setting out conditions that the parties must meet. However, British Columbia courts have a residual discretion because that province’s statute states that “the court may grant leave” if those conditions are met. An Alberta court, on the other hand, is told that it “shall grant [leave] only if it is satisfied” those conditions are met. The use of the word “shall” suggests an absence of residual discretion in Alberta, but the House of Lords in Pioneer Shipping Ltd. et al. v. B.T.P. Dioxide Ltd., [198013 All ER 117 (CA), aff'd [1981] 2 All ER 1030 at 1034 (HL), held that a very similar leave to appeal provision in the English legislation did not affect the general discretionary nature of the granting of leave (see also John J. Chapman, “Judicial scrutiny of domestic commercial arbitral awards”(1995) 74 Canadian Bar Review 401 at 412).

Finally, the conditions for granting leave in Alberta are narrowly focused on only the parties. It is only the importance of the dispute to the parties and whether their rights will be significantly affected by the appeal that counts. The courts are seen as dispute resolvers only, a continuation of the role of private arbitration. In British Columbia, appeals from arbitration may have a wider focus, beyond that of the parties only. The appeal can be important to the parties or to a group that the appealing party is a member of, or to the general public. Some Alberta courts have read into the Alberta appeal provisions the requirement that the appeal be in the public interest. See, for example, my 2011 post “Leave to Appeal an Arbitration Award: Is There a Public Interest Requirement?” (The Alberta Law Reform Institute has examined the public interest requirement in Arbitration Act: Stay and Appeal Issues, 2013 (Final Report 103) in the context of re-thinking appeals from arbitration awards and it has recommended that section 44(2) be repealed in its entirety and that there be no appeals without the parties’ agreement (para 138)).

So what do these similarities and differences in the British Columbia and Alberta legislation amount to when it comes to applying Sattva in this province?

Most importantly, the Court’s precedent-setting decision that contractual interpretation issues are questions of mixed fact and law will apply to restrict potential appeals from arbitration awards in Alberta. Contract interpretation questions are prominent in commercial arbitrations and, because they are no longer questions of law and leave to appeal is only available for questions of law, there should be fewer requests for leave to appeal and fewer leaves to appeal granted. It is not quite the abolishment of all appeals with leave that the Alberta Law Reform Institute recommended, but it is a big step in that direction. Members of the legal profession and the public will know less and less about arbitration as cases such as Sattva will simply be decided, privately and confidentially, by decision-makers chosen by the parties to the dispute.

The Court’s interpretation of section 32(1)(a) — the importance of the result of the arbitration to the parties justifies the intervention of the court and the determination of the point of law may prevent a miscarriage of justice — will probably be persuasive in Alberta. Section 31(1)(a) is the closest the British Columbia legislation comes to Alberta’s party-focused conditions for granting leave. Recall that section 44(2) of the Alberta legislation requires that the importance to the parties of the matters at stake in the arbitration justifies an appeal and that determination of the question of law at issue will significantly affect the rights of the parties. This is similar in substance, if not form, to the British Columbia provision. Therefore the Court’s interpretation that what was required was that “an alleged legal error must pertain to a material issue in the dispute which, if decided differently, would affect the result of the case” (para 70) is probably an accurate description of what section 44(2) requires.

The Court’s narrowing of the discretion of the court hearing the leave to appeal application is probably not relevant to Alberta. It is not clear that courts in Alberta have a residual discretion to decline to grant leave if the conditions in section 44(2) are met. Section 44(2) does not say that courts must grant leave if the conditions are met, but it does say they “shall” grant leave only if the conditions are met. Thus it appears that Alberta courts cannot take “the parties’ conduct, the existence of alternative remedies, and any undue delay” (para 87) into account in deciding leave application under section 44(2).

The Court’s final point, namely, that the standard of review will almost always be reasonableness, barring a constitutional question or a question of law of central importance to the legal system as a whole and outside the adjudicator’s expertise, will apply in Alberta. It is a general point about arbitration and the types of issues that are generally argued and determined in arbitration.

In summary, the principles of arbitration law decided by the Supreme Court in Sattva should be highly persuasive in Alberta, despite the differences in the arbitration statutes in British Columbia and Alberta.

What happens when A sells B a working interest in the thermal or enhanced production from an oil and gas property and A or its successors in interest continue with primary production?

Mon, 08/18/2014 - 12:00pm

By: Nigel Bankes

Case Considered: IFP Technologies (Canada) v Encana Midstream and Marketing, 2014 ABQB 470

What happens when A sells B a working interest in the thermal or enhanced production from an oil and gas property and A or its successors in interest continue with primary production? This was the issue at the heart of this decision. The answer is that B gets shafted; B should have taken better steps to protect itself rather than simply assuming that all future production from the property would take the form of enhanced or thermal production.

In the course of his lengthy 73 page judgement Chief Justice Neil Wittmann (acting in place of Justice Ron Stevens (deceased)) addressed a number of questions of oil and gas law which will be of interest to the energy bar including the following: (1) What property interest did IFP acquire? (2) What is the test for determining whether a working interest owner has reasonable grounds for refusing consent to an assignment of shared interest lands under the 1990 CAPL Operating Procedure? (3) What is the legal position where a working interest purports to withhold consent and the Court subsequently determines that the withholding of consent was unreasonable? (4) Did the development of the property through primary production techniques substantially nullify the benefit for which IFP (B) had bargained so as to amount to a breach of contract? (5) Assuming that there was a breach of contract how should damages be assessed? (6) Assuming liability should any claim for damages be capped by a contractual agreement between the parties?

The facts and the agreements between the parties?

IFP (a wholly owned subsidiary of IFP Energies Nouvelles of France) had expertise and technical information in relation to the drilling, placement and completion of horizontal wells. Beginning in the late 1980s IFP entered into a series of agreements with CS Resources, a pioneer in the use of horizontal wells for the development of heavy oil resources. As part of the first series of these agreements CS Resources granted IFP a 3% gross overriding royalty (GOR) on all CS lands on which IFP’s technology was applied. PanCanadian (PCR) acquired CS Resources in 1997.  IFP and PCR eventually concluded that the GOR model was inappropriate and agreed to replace it with a working interest model. That agreement was recorded in an MOU of July 1998 and an Asset Exchange Agreement (AEA) of October 1998 to which were scheduled a joint operating agreement (JOA) and its appended operating procedure, which was an amended version of the CAPL 1990 Operating Procedure.  This second set of agreements covered both the original CS Resources lands as well as other lands rolled in to the deal by PCR including properties referred to as the Eyehill Creek Assets. At the time of the AEA there were already 222 conventional wells on these lands.

Under the AEA IFP was to acquire a 20% working interest in the PCR lands including the Eyehill Creek Assets. The granting language of the AEA provided as follows:

[67] PCR hereby agrees to sell, assign, transfer, convey and set over to IFP, and IFP hereby agrees to purchase from PCR, all of the right, title, estate and interest of PCR (whether absolute or contingent, legal or beneficial) in and to the PCR assets, … all subject to an in accordance with the terms of this Agreement.

(Emphasis is CJ Wittmann’s)

The idea that IFP’s interests were actually limited to thermal and enhanced production first seems to have been introduced in the terms of the JOA which was scheduled to the AEA. Clause 4(c) provided (at para 92):

4(c) It is specifically agreed and understood by the parties that the working interests of the parties as described in Clause 5 of this Agreement relate exclusively to thermal or other enhanced recovery schemes and projects which may be applicable in respect of the petroleum substances found within or under the Joint Lands and the Title Documents. Unless specifically agreed to in writing, IFP will have no interest and will bear no cost and will derive no benefit from the recovery of petroleum substances by primary recovery methods from any of the rights otherwise described as part of the Joint Lands or the Title Documents.

(Emphasis is CJ Wittmann’s)

Other provisions of the JOA, including the definition of working interest and the nature of the parties’ participating interests, all simply referred to the interests of the respective parties in the lands without further qualification by reference to the nature of the production process.

Under the attached CAPL operating procedure the parties had elected the right of first refusal option (ROFR) under Article 24 (at para 110) and the agreement seems to have contained the standard provisions on independent operations (with some amendments) with a 400% penalty (at para 107). Another element of the JOA was a series of clauses that relieved IFP of any responsibility for the abandonment of the conventional (primary production) wells on the Eyehill Creek property (at para 33).

By the late 1990s PCR was concerned about its ability to hold on to the Eyehill Creek lands and was focusing on developing other assets such as its Christina Lake property.  One of PCR’s Eyehill leases had expired and in other cases Alberta Energy had issued notices on continued leases requiring PCR to establish the productivity of the properties. Oil prices were depressed and PCR had shelved any idea of introducing a thermal recovery operation at Eyehill. Given these concerns and concerns as to the abandonment liabilities associated with its existing wells, PCR was receptive to proposals to removing itself from the property. In 2001 PCR executed a letter agreement with Wiser which was a form of farmout agreement (ultimately formalized as an Abandonment, Reclamation and Option Agreement (ARO)) pursuant to which Wiser would earn PCR’s working interest in the Eyehill Creek lands by “dealing with” the existing 222 wells by abandonment and reclamation, by re-working them or by putting them on production. It was clear that Wiser was only interested in the primary production possibilities from these lands.

PCR gave IFP the ROFR notice to which it was entitled in April 2001. At about the same time Wiser also sought (unsuccessfully) to clarify with IFP that IFP’s working interest was confined to enhanced and thermal recovery operations. IFP declined to exercise its ROFR but did withhold consent to the disposition on the grounds that Wiser “had no technical capability or intent to pursue thermal or other enhanced recovery” (at para 52; see also para 167). PCR proceeded to execute the ARO. Wiser protected itself through an indemnity agreement with PCR. Wiser was never novated into the AEA and related agreements.  Wiser commenced the operations contemplated by the ARO and earned its interest. Wiser never informed or consulted IFP as to the nature of those operations. Canadian Forest acquired Wiser’s interests in 2004. All of the operations conducted by Wiser and Canadian Forest were primary production operations; none involved enhanced or thermal recovery.

On the basis of these facts IFP alleged that the Wiser farmout (the ARO) was a breach of contract and sought damages. PCP took the view that IFP had unreasonably withheld its consent to the proposed agreement.

What property interest did IFP acquire?

I think that there are two possible interpretations of what IFP acquired. One interpretation (which I will refer to as the property-limited-by-contract interpretation) is that IFP acquired an undivided interest as a tenant in common of the relevant Crown leases and other assets (subject to some contractual limitations on its precise rights in relation to those assets). A second interpretation (which I will refer to as the property interpretation) would hold that IFP acquired something in the nature of a working interest in production from the lands resulting from thermal or other enhanced recovery techniques.  There are pros and cons to each of these interpretations.

The principal argument in favour of the property-limited-by-contract interpretation is that that it is the natural interpretation of the granting words used in the dominant agreement, the AEA. It also has the advantage that it accords IFP a legally coherent and cognizable interest in the property. We know what the basic rights of a tenant in common are. The contrary argument is that this classification does not seem to be consistent with the overall intentions of the parties which suggested that IFP’s rights prima facie did not extend to primary production. But the best way to respect that intention is to conclude that the property rights of IFO as a tenant in common were limited by the terms of the other contractual arrangements between them, including the key provision in the JOA referred to above.

The principal argument in favour of the property interpretation is that it delivers a result that seems to comport with the overall result intended by the parties reading all of the agreements together and the commercial context for those agreements. The principal knock against this interpretation is that it fails to respect the dominant conveyancing language of the AEA and as a result delivers an interest which is unrecognizable in terms of property law. It is one thing to have an undivided interest which is confined to a particular formation or formations; or to have an undivided interest in a particular substance; but we create a whole new layer of complexity when we admit of the possibility that ownership of an interest in land varies with the nature of production from those lands. Not only is this complex but it seems to be inconsistent with the royalty-as-interest-in-land cases culminating in Bank of Montreal v Dynex Petroleum Ltd, 2002 SCC 7. If an interest in the proceeds of production cannot give an interest in land how can a party have a tenancy in common (not just any old interest in land, but an undivided interest) in a Crown lease that is contingent on the mode of production of the leased substances?

How was this issue resolved here? Chief Justice Wittmann seems to suggest that both the plaintiff and the defendant adopted some version of the property-limited-by-contract approach but the Chief Justice himself preferred some version of the property approach:

[97] I find that IFP’s working interest pursuant to these agreements has always been limited to thermal and other enhanced recovery methods. I find the AEA did not grant broad rights that were subsequently reduced or modified by the JOA, as assumed by both the Plaintiff and the Defendants. The AEA does not define the term working interest. The Preamble to the AEA states, however, that the ownership of working interests is subject to and in accordance with the terms and conditions of the JOA. Furthermore, the JOA is incorporated by reference into the AEA as though it were contained in the body of the AEA. As such, the definition of working interest in the JOA is incorporated by reference into the AEA.

See also para 194 where the Chief Justice comments further on the relationship that the parties have created.

But whatever interpretation is adopted it is still necessary to work through the applicability of the operating procedure to primary production. We don’t have the complete story from the judgement and in particular we do not know the full extent to which the parties modified the CAPL 1990 form, but one would anticipate that significant changes would be required to make it work in these circumstances. Consider, however, what we do know. We know (see para 54) that Wiser carried out operations on the lands once it had acquired its interest in the property and we know that it did not inform IFP about those operations. We can infer from this that Wiser was not in the habit of sending IFP AFE (authorizations for expenditure) notices (which passes without comment in the judgement). Yet on the other hand the Court and the parties assume the applicability of the independent operations clause (modified as discussed at paras 105-107) with the result that Chief Justice Wittmann concludes that IFP might have been able to trigger the clause – although as a matter of practice it lacked both the capital and the operational expertise to be able to do so (at para 197). But even aside from this practical problem facing IFP, it would be extremely difficult legally for IFP to propose an effective independent operation where there were already licensed wells for the relevant drilling spacing units.

The complexities of determining the applicability of various clauses of the CAPL procedure (absent an express statement as to (in)applicability) seem legion. What about the applicability of the CAPL provisions dealing with access to information? Was IFP entitled to information about primary production from the lands (referred to at para 176)? What about Article XI dealing with the surrender of joint lands (referred to at para 221)?

The difficulties were also evident with respect to Article 24, the ROFR/consent provision of the procedure. Given Chief Justice Wittmann’s conclusions as to just what it was that IFP had obtained (i.e. a working interest in only thermal and enhanced production) there was a certain logic to PCR’s position (at para 140) that the transfer to Wiser should not trigger Article 24 since Wiser was only interested in primary production. The difficulty with that argument however was that whatever Wiser’s intentions with respect to what it would produce (and how), Wiser was clearly acquiring PCR’s entire interest in the property. Thus Chief Justice Wittmann is surely correct in concluding (at paras 141-145) that the Wiser transaction did trigger Article 24. The question would have been more difficult had PCP retained its rights to thermal and enhanced production.

What is the test for determining whether a working interest owner has reasonable grounds for refusing consent to an assignment of shared interest lands under the 1990 CAPL Operating Procedure?

The ROFR provision of the 1990 CAPL afford each working interest owner (WIO) two independent rights: the ROFR right itself and the right to refuse consent to the proposed transfer even where the WIO will not exercise the ROFR.

2401B(e)  In the event that the working interest described in the disposition notice is not disposed of to one or more of the offerees pursuant to the preceding Subclause, the disposition to the proposed assignee shall be subject to the consent of the offerees. Such consent shall not be unreasonably withheld, and it shall be reasonable for an offeree to withhold its consent to the disposition if it reasonably believes that the disposition would be likely to have a material adverse effect on it, its working interest or operations to be conducted hereunder, including, without limiting the generality of all or any part of the foregoing, a reasonable belief that the proposed assignee does not have the financial capability to meet prospective obligations arising out of this Operating Procedure. …

(Emphasis is CJ Wittmann’s)

This gives rise to two questions. The first is really a methodological question – how should the Court go about analyzing such a question. And the second is that of how to apply the preferred approach to the facts at hand. As for the methodology, both counsel and the Court (at para 152) decided to rely on case law dealing with the unreasonable withholding of consent in the context of the landlord and tenant relationship. There might be some doubts as to the applicability of this body of law in this setting and thus it is useful to have the Court affirm its relevance. From this body of law the Chief Justice derived the following principles:

[153]  The burden of proof is on the party asserting consent was unreasonably withheld: Sundance Investment Corporation Ltd v Richfield Properties Limited (1983), 41 AR 231 at para. 23 (CA).

[154]    The party whose consent is required is entitled to base its decision on its own interests alone: Community Drug Marts P & S Inc, Estate of v William Schwartz, Construction Co Ltd, 31 AR 466 at para 41, (QB), aff’d [1981] AJ No 537.

[155]    Whether a person has acted reasonably in withholding consent depends on all the factual circumstances:  Exxonmobil Canada Energy v Novagas Canada Ltd, 2002 ABQB 455 at para 49. The question is not whether a reasonable person might have given consent, but whether a reasonable person could have withheld consent in the circumstances: 1455202 Ontario Inc v Welbow Holdings Ltd, [2003] OJ No 1785 at para 9 (ONSC) (“Welbow”). In Exxonmobil, Park J reviewed the evidence on an objective basis to determine whether in the circumstances a reasonable person would have refused to consent to the assignment.

[156]    A party must not refuse consent where such refusal is calculated to achieve a collateral purpose, or benefit, not contemplated by the original contract: Welbow at para 9.

[157]    Proceeding with an assignment in the face of a reasonable refusal to consent is a clear breach of a negative covenant: Exxonmobil at para 51.

[158]    The court should not defer to the party withholding consent, but must assess the reasons for withholding consent and consider whether a reasonable person in similar circumstances would have made the same decision. The court should consider the purpose of the consent clause and the meaning and benefit it was intended to confer.

Notably absent from this list is any reference to the venerable decision of the English Court of Appeal in Houlder Brothers v Gibbs, [1925] 1 Ch 575, which stands for the proposition that a lessor will be able to withhold consent on grounds related to the personality of the proposed assignee or the use and occupation that the proposed assignee will make of the leased premises. Admittedly it is very difficult to reconcile Houlder Brothers with the majority decision of Alberta’s Court of Appeal in Sundance, but recall that in Sundance the majority was clearly of the view that a lessor had good grounds to object to any assignment that prejudiced the lessor’s financial interest. I have never been very persuaded by that approach and much prefer Justice Harradance’s dissenting judgement but in this case both Houlder and the majority judgement in Sundance seemed to offer some comfort to IFP.

Indeed, if one looks simply to the outcome of the transfer in this case it look like a case in which IFP should be able to withhold consent. After all, if IFP failed to forestall the transfer it was going to be forced into a joint venture with a party that had the announced interest of exploiting the property exclusively for its primary production potential. Not only would that exclude IFP from the opportunity to take its 20% share of production, it would also prejudice the economics and perhaps physical feasibility of future enhanced or thermal recovery operations at the site. But for Chief Justice Wittmann this was an oversimplification. He concluded that IFP’s withholding of consent was unreasonable.

Ultimately I think that the principal reason for this conclusion is that as a matter of law IFP is no worse off after the Wiser transaction than it was before the transaction. This is because PCR was under no legal obligation to develop the thermal and enhanced recovery potential of the lands. IFP had failed to contract for that obligation.  One may question how consistent this is with the landlord and tenant cases which I think clearly allow the landlord to use the right to withhold consent as a means of ensuring that the property is not used for certain purposes even though the landlord had not specifically contracted against those uses in the lease: Houlder Brothers and Sundance both support that proposition.

Perhaps more convincing is Chief Justice Wittmann’s overall assessment of (un)reasonableness in light of the dire circumstances facing PCR (and therefore ultimately IFP itself). Essentially PCR was sitting on a dying property in the form of a set of leases (although PCR did hold the freehold mineral title to some of the lands) that were going to expire or be cancelled unless somebody did some work on the property (and PCR certainly had no obligation to do that). Seen in this light the transfer to Wiser was a means of saving the properties and saving IFP’s interest in those properties even if it might have prejudiced the adoption of thermal and enhanced recovery in the future. In other words, better the chance of the continuing possibility of future thermal and enhanced recovery (however remote) than the inevitable (and relatively immediate) loss of the properties. But if one takes this broad view of reasonableness then it might also be necessary to consider the extent to which the dire circumstances in which PCR found itself were inevitable or whether they were of PCR’s own making.

What is the legal position where a working interest purports to withhold consent and the Court subsequently determines that the withholding of consent was unreasonable?

If a tenant assigns a lease in breach of the covenant not to assign or sublet without the landlord’s consent (such consent not to be unreasonably withheld) the assignment or sublease is not invalid or void but the tenant is in breach of its covenant and the landlord will typically have reserved a right of re-entry for breach. Similarly, if the landlord withholds consent and the tenant believes the withholding to be unreasonable the tenant may elect to proceed knowing that if it can establish that the landlord’s behavior is unreasonable it will not be in breach of its covenant. This is a high risk course of action since in the case of a lease the penalty for being wrong may be the loss of the lease. As a result, the assignee may well, as here, demand an indemnity. High risk it may be but it is a more expeditious way of proceeding than the alternative which is to apply for a declaration as to the unreasonableness of any withholding of consent (and note that under the CAPL the arbitration provisions of Article 24 apply to valuation issues in package deals; they do not apply to the consent issue).

The issue is a bit more complicated in the context of CAPL because of the novation provisions of the agreement – modified in this case and universally by the terms of the CAPL Assignment Procedure. These provisions are designed to provide for deemed novation in certain circumstances but the provisions can only be triggered if the parties are in compliance with the consent provisions.

In this case Chief Justice Wittmann concluded that the logic of all of this was applicable to the joint operating context and thus: (1) PCR was not in breach of the covenant not to assign without consent because consent was withheld unreasonably, (2) the deemed novation provisions were not precluded from applying by the absence of consent, and (3) therefore Wiser had been novated into the relevant agreements.

The reader may be wondering where this argument was going and who was on what side of it. The issue had been raised by IFP. IFP wanted to argue that if Wiser had not been novated into the JOA the provisions in the JOA that limited IFP’s interest to an interest in thermal or enhanced recovery could not be enforced against IFP – IFP could then be taken to have an unqualified 20% undivided interest in the property. And on that basis IFP sought an accounting of its share of production relying on the Statute of Anne, 4 Anne c 16, s 27 (UK). Chief Justice Wittmann concluded (at paras 402- 403) that his earlier findings as to the limited nature of IFP’s interest and his conclusion on the novation argument just referred to were a complete answer to the claim for an accounting.

Did the development of the property through primary production techniques substantially nullify the benefit for which IFP (B) had bargained so as to amount to a breach of contract?

It seems to me that Chief Justice Wittmann dealt with this issue in two parts of his judgement, first at paras 199-212 under the heading “4. What is the relevance of the reasonable expectations of the parties?’ and then later at paras 220-270 under the heading “6. Has the opportunity to pursue a thermal or other enhanced recovery project at Eyehill Creek been destroyed or damaged?”  In framing the issue in terms of substantial nullification rather than adopting the Chief Justice’s headings I am drawing on Justice Kerans’ judgement in the Court of Appeal in Mesa Operating Ltd Partnership v Amoco Resources (1994), 149 AR 187 (which the Chief Justice refers to at paras 199-201). I think that the Mesa case and the substantial nullification test referred to in that decision provide an appropriate umbrella for the consideration of these two headings in part because there is no discussion of any applicable law under heading (6) in the chief justice’s judgement. Thus it seems best to bring the “destroyed or damaged” framing of heading (6) under the Mesa umbrella.

In Mesa, Mesa held a GOR in half a section of lands and argued that Amoco breached its contractual obligations to Mesa when it carried out an administrative pooling of its lands on an acreage basis rather than on a reserves basis thereby effectively diluting Mesa’s royalty entitlement. Amoco had the power to pool under the terms of the GOR agreement and thus the question was whether it had abused its discretion in the manner in which it went about exercising that power. The Court of Appeal concluded that this was a case in which pooling should have taken place on a reserves basis largely because it was able to say, considering the traditions and practices of the industry, that it was well established that “an operator pools on a reserves basis if the geographical data clearly shows the boundaries of the reservoir, and those boundaries are significantly at variance with the size of the corresponding surface parcels …”. Given the unusual nature of the split rights in this case it was clearly going to be difficult for IFP to establish an analogous body of practice to support its contentions in this case.

Chief Justice Wittmann concluded that IFP could not make out its case under either of these two headings. IFP had not bargained for a prohibition on primary production (at para 212) (and thus that benefit was not in the contemplation of both parties and had not been nullified) and while there was much evidence that it would be more difficult and more expensive to introduce a thermal or enhanced recovery operation into a field that had been drilled out and depleted through conventional recovery measures and conventional cementing jobs, such an operation would not be impossible (at paras 267 – 268). In so concluding the Chief Justice establishes that Mesa sets a very high threshold. The application of the test does seem justified in this case because the parties clearly contemplated some continuing primary production, and, as the Court notes at para 195, given that, some level of conflict between those who own all the rights and those who only own some rights (the right to enhanced or thermal production) is inevitable.

Assuming that there was a breach of contract how should damages be assessed?

Although Chief Justice Wittmann concluded that PCR was not liable to IFP he did go on and consider whether IFP had been able to establish that it had suffered any damages. The Chief Justice posed three questions: (1) Was the claim of lost opportunity to develop the thermal and enhanced recovery potential of the property real or fanciful? (2) If real what was the value of the opportunity? (3) What was the likelihood that IFP would have been able to realize this opportunity and what discounting factor should be applied?

Chief Justice Wittmann concluded (at paras 284- 285) that the claim of lost opportunity was not merely fanciful. PCR disposed of the Eyehill Creek property for strategic reasons not because it believed that that the property had no potential for thermal development. He was less sympathetic to the plaintiff on the other two questions concluding (at para 364) that the plaintiff had been unable to establish any value for its lost opportunity and concluding further that there was zero chance that PCR would have initiated a thermal recovery operation in the absence of a farmout because of the poor economics and IFP would have been unable to initiate such an operation itself. I have not dug too deeply into these sections of the judgement but they seem very much to emphasise the economics of a thermal recovery project based upon oil prices at the time of the farmout. The rationale for focusing on the price environment at that time is that PCR would not have been able to hold on to the properties (see paras 377-378) and wait for prices to improve.

Assuming liability should any claim for damages be capped by a contractual agreement between the parties?

Article 9 of the AEA provided that in no event should PCR’s liability to IFP exceed the value of the PCR assets. The parties assigned a value of $16 million to those assets; IFP’s claim for damages was for $45 million. Chief Justice Wittmann commented as follows:

[405] On its face, a limitation of damages clause is legitimate and enforceable. IFP and PCR are sophisticated business entities who negotiated the AEA with the assistance of legal counsel. There is no indication of unconscionability or oppression at the time the contract was negotiated. There are also no public policy reasons to ignore the limitation clause.

[406]…. Given the language of the contract, IFP’s claim for $45 million in damages was untenable.

Such limitation of damages clauses are common in purchase and sale agreements for oil and gas properties and confirmation of their enforceability will be welcomed.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca 

Follow us on Twitter @ABlawg

Rectification of Conservation Easement Agreements on the Basis of a Mistake in Integration: Underlying Principles

Mon, 08/11/2014 - 10:00am

By: Evaristus Oshionebo

PPF Version: Rectification of Conservation Easement Agreements on the Basis of a Mistake in Integration: Underlying Principles

Case Commented on: Nature Conservancy of Canada v Waterton Land Trust Ltd, 2014 ABQB 303

This case raises a myriad of legal issues covering disparate areas of the law. For the purpose of this post, only those facts relevant to the issue of rectification of the conservation easement agreement will be discussed. Other aspects of the facts can be found in Jonnette Watson Hamilton’s earlier post here.

Facts

The Nature Conservancy of Canada (NCC), a private non-profit corporation, owned the Penny Ranch located in the eastern slopes of the Rocky Mountain. The ranch is strategically important for the movement of wildlife in Alberta as it is within the migratory corridors of several species of wildlife. Thus the NCC was interested in preserving the ranch as a migratory corridor for wildlife.

Once it became apparent to the NCC that it might soon be selling the Penny Ranch, the NCC decided to register a conservation easement on its title to the land. A conservation easement is a written agreement between a registered owner of land and a conservation organization or government agency designed to facilitate the protection, conservation and enhancement of the environment. See Alberta Land Stewardship Act, SA 2009, c A-26.8, s. 29. The NCC intended that the terms of the conservation easement would later be revised to suit the purchaser of the ranch, provided that any such revision did not compromise the NCC’s ability to conserve the ranch.

The NCC believed that, as owner of the Penny Ranch, it could not hold a conservation easement on the ranch in its own name. Thus, the NCC enlisted the assistance of the Alberta Conservation Association (ACA), a charitable organization dedicated to the conservation of nature. The ACA agreed to hold a placeholder conservation easement with regard to the Penny Ranch. On May 28, 2003, the NCC (as grantor) and the ACA (as grantee) entered into a conservation easement agreement regarding the Penny Ranch. The conservation easement agreement was duly registered on the NCC’s title to the ranch. The NCC and the ACA understood that should the NCC sell the ranch to a third party in the future, the ACA would transfer or assign their rights under the conservation easement agreement to the NCC.

The conservation easement agreement contained a number of restrictions on use of the Penny Ranch including restrictions on fencing of the ranch. The agreement provided in part that:

The Grantor may maintain, replace and repair the fences, roads, buildings, and other improvements (Facilities) located on the Property as of the date of this easement. The Facilities are to be maintained, replaced or repaired, each at its original size and in its same location. If any or all of such Facilities are removed or destroyed, the Grantor may replace them with similar structures of the same size in the same location.

In 2003, the NCC decided to sell the Penny Ranch subject to a conservation easement to ensure that the land is conserved in perpetuity. To this end, the NCC approached the Defendant, Thomas Olson, to purchase the ranch. Mr. Olson, an avid conservationist, operated bison ranches in Western Canada. Olson was interested in purchasing the Penny Ranch because he thought it would be suitable for his bison ranching business.

The NCC’s Director of Land Conservation, Margaret Green, negotiated the sale of the Penny Ranch on behalf of the NCC. Green had the full authority of the NCC to negotiate and close the sale of the ranch. The negotiations between Olson and Green were amicable and constructive given that both the NCC and Olson were mutually interested in the conservation of nature. In the end, Green and Olson agreed orally on the terms of sale and purchase of the ranch. They also agreed orally to amend the conservation easement agreement previously registered on the NCC’s title to the ranch. Olson testified that, in order to cater to his bison ranching business, the parties orally agreed on specific amendments to the conservation easement agreement including the fence height restrictions. This oral agreement is referred to by the court as the “Agreed Fence Height Restriction”.

Olson then instructed his solicitor to prepare a formal offer to purchase the property in accordance with the terms of his oral agreement with Green. Olson’s solicitor prepared and sent the Offer to Purchase to the NCC. Attached to and forming part of the Offer to Purchase was a document purportedly containing the amendments which Green and Olson had agreed to make to the conservation easement agreement. However, due to an error on the part of Olson’s solicitor, this document (referred to by the court as the ‘Conservation Easement Amending Agreement’) did not correctly reflect the amendments orally agreed to between Olson and Green. Rather, the document included the fencing restrictions (particularly fence height restriction) in the initial conservation easement agreement. More specifically, the document contained the following provisions:

Property Management Principles

8.         Section 1.01 of Schedule B of the Agreement is hereby deleted and replaced with the following:

…………

1.1      The Grantor may maintain, replace and repair the fences, roads, buildings, and other improvements located on the Property as of the date of this easement. The fences and roads are to be maintained, replaced or repaired, each at its original size and in its same location. If any or all of the buildings are removed or destroyed, the Grantor may replace them with structures of a similar purpose at or near the same location within the existing 5 acre home site. Any building construction shall require the prior notice to the Grantee.

 (Emphasis added by the court)

The ‘Conservation Easement Amending Agreement’ was to be entered into between the NCC, as owner of the Penny Ranch and grantor of the initial conservation easement, and the ACA as grantee of the initial conservation easement. However, the terms of the ‘Conservation Easement Amending Agreement’ were determined by the NCC and Olson as part of the sale and purchase of the Penny Ranch.

The Offer to Purchase, along with the documents attached to it, was accepted by the NCC and the sale of the Penny Ranch to Olson was closed on the basis of the terms of the Offer to Purchase. Subsequently, the NCC and the ACA executed the ‘Conservation Easement Amending Agreement’.

Soon after purchasing the Penny Ranch, Olson decided to replace the fencing on the perimeter of the ranch with a new fence. In September 2004, Olson commenced construction of a new fence which he believed would be more effective in restraining and keeping his bison. The NCC received several phone calls from neighbouring landowners complaining about the new fence. As a result the NCC asked Olson to change the fence. The NCC took the view that the height of the new fence breached the terms of the Conservation Easement Amending Agreement which, the NCC alleged, is the document attached to Olson’s Offer to Purchase the ranch. The NCC also believed that the new fence would hinder wildlife migration.

The NCC brought this action against Olson seeking enforcement of the terms of the Conservation Easement Amending Agreement. The NCC sought an order compelling Olson to modify his new fence. Olson counter-claimed against the NCC for an order rectifying the Conservation Easement Amending Agreement so that it accords with the oral agreement reached by Olson and the NCC. Olson contended that, in the course of drafting the Offer to Purchase, there was a mistake in integrating the terms of the Conservation Easement Amending Agreement.

In September 2005, the NCC and the ACA signed an assignment agreement under which the ACA transferred their rights under conservation easement agreement to the NCC, as originally contemplated by the parties.

The Decision

Terms of the Oral Agreement

The Court of Queen’s Bench (Justice P.R. Jeffrey) accepted Olson’s testimony regarding the oral agreement he reached with the NCC, finding in the process that “Olson’s recollection and testimony on these negotiations was both clear and credible” (para 109). The court held that Olson and the NCC orally agreed on the sort of fencing that was necessary for the adequate containment of bison on the Penny Ranch. In the words of the court,

[114] I find that the NCC regarded Olson’s proposed changes to the Initial ACA CE in respect of fencing acceptable. I find that Green on behalf of the NCC readily agreed with Olson that his desired changes to the Initial Fence Height Restriction would get included in the amendment to the Initial ACA CE and, thereby, become a part of the final NCC conservation easement.

The court concluded (at paras 126 & 327) that the parties had an oral agreement (characterized by the court as the “Agreed Fence Height Restriction”) as follows:

 

The Grantor may maintain, replace and repair the fences, roads, buildings, and other improvements located on the Property. If doing so with fences or roads, they are to be maintained, replaced or repaired at or near the existing ones. The Grantor may not build fences or roads in areas where none exists without the Grantee’s permission. The building of wildlife-proof fences is not permitted, except in localized areas as needed to control or prevent wildlife damage to haystacks, stored forage or domestic gardens. If any or all of the buildings are removed or destroyed, the Grantor may replace them with structures of a similar purpose at or near the same location within the existing 5 acre home site. Any building construction shall require the prior notice to the Grantee.

 

The court held further that this oral agreement, that is, the “Agreed Fence Height Restriction”, did not preclude Olson from constructing the new fence.

 

Although there is no documentary evidence indicating the precise terms of the oral agreement that the parties reached on fencing of the Penny Ranch, there is strong circumstantial evidence indicating that the parties did reach an oral agreement on fencing of the property. Circumstantial evidence also points to the fact that the parties agreed that the fence height restriction in the initial conservation easement agreement would not govern Olson’s purchase and use of the ranch. For example, the NCC was aware in the course of negotiations that Olson intended to use the property for his bison ranching business. They were also aware that the fence height restriction in the initial conservation easement agreement was a potential deal breaker for Olson because the restriction would render the Penny Ranch unsuitable for bison ranching (paras 111-113). As the court held (at para 111),

 

… (1) absent certainty on the ability to install adequate fencing for bison ranching on the Property, Olson would not have proceeded with the purchase of the Penny Ranch; and (2) if the NCC had any concerns with Olson’s wishes on perimeter fencing, Green would have raised them and remembered at trial any ensuing discussions, because of their criticality to Olson’s purpose for acquiring the Property. They would have been every bit as, if not more, memorable as their discussions on additional buildings.

 

Further circumstantial evidence is the fact that Olson’s solicitor, pursuant to Olson’s instructions, sent a revised and modified version of the conservation easement agreement to the NCC, wherein Olson’s solicitor deleted the phrase “each at its original size and in its same location” (para 125). This phrase formed an integral part of the initial conservation easement agreement between the NCC and the ACA. Olson’s solicitor apparently deleted the phrase from the agreement in order to ensure that it accorded with the parties’ prior oral agreement on fencing.

 

Moreover, the NCC’s conduct prior to and after the sale of the ranch to Olson shows that the height of the fence was not a matter of particular concern to the NCC at the time of sale of the ranch to Olson. For example, the NCC did not include the phrase “at its original size and in its same location” in either the Grazing Lease Agreements it signed with other parties prior to the sale to Olson or its grazing lease with Olson. In addition, the NCC’s Grazing Plan for the Penny Ranch was silent on fencing, while the Penny Ranch Baseline Report prepared by an independent expert on behalf of the NCC did not address the matter.

 

In the face of these documents, it “is simply too implausible” that the parties would have intended the fence height restriction in the initial conservation easement agreement to govern the sale of the property to Olson. According to the court (at para 330), “[t]he approaches to fencing in these contemporaneous documents are also further evidence that fence size was not on the NCC’s radar screen” at the time the property was sold to Olson.

 

Unfortunately, the oral agreement did not get correctly reflected in the Conservation Easement Amending Agreement that the parties signed (para 120).

 

Should the Conservation Easement Amending Agreement be Rectified?

 

Having found that the parties’ oral agreement was not correctly reflected in the written Conservation Easement Amending Agreement, the court then proceeded to determine whether the written agreement should be rectified to accord with the prior oral agreement. A mistake in integration of contract may be rectified by the court for the sole purpose of restoring the parties to their original agreement (Performance Industries Ltd. v. Sylvan Lake Golf & Tennis Club Ltd., [2002] 1 SCR 678 at para 31). However, not all types of mistakes can support rectification of a contract. Rectification is granted only in instances involving mutual mistake and unilateral mistake.

 

Even then, a party seeking rectification of a contract on the basis of mutual mistake or unilateral mistake must provide “convincing proof” of their case (Performance Industries at para 41). The requirement of ‘convincing proof’ imposes a burden of proof higher than the burden required in civil cases which is proof on balance of probabilities. While the ‘convincing proof’ standard falls short of the criminal standard (that is, proof beyond a reasonable doubt), it “goes beyond the sort of proof that only reluctantly and with hesitation scrapes over the low end of the civil ‘more probable than not’ standard” (Performance Industries at para 41).

 

Mutual Mistake

 

On the issue of mutual mistake, the court held (at para 345) that “there was a mutual mistake between the parties on the fencing provision”. A mutual mistake occurs where both parties share a mistaken assumption regarding some important matter underlying their contract. In such instances, “the parties reach agreement on the terms of their contract but share an error with respect to some important contextual circumstance that has motivated one or both of the parties to enter the agreement” (John D. McCamus, The Law of Contracts, 2nd ed (Irwin Law, 2012) at 556).

 

In the instant case, it appears that both parties had a mistaken assumption that the Conservation Easement Amending Agreement correctly reflected their prior oral agreement. As the court observed (at para 352), “The mistake here was not in the bargain struck between the parties on fencing but in the failure to include correctly in the written agreement that bargain they reached”.

The court articulated (at para 334) the requirements for rectification of a contract on the basis of mutual mistake as follows:

 

The party seeking rectification must establish on a standard of convincing proof:

 

(i) the existence and nature of a common intention by the parties prior to the making of the document or instrument alleged to be deficient; (ii) that this common intention remained unchanged at the date that the document or instrument was made; and (iii) that the challenged document or instrument, by mistake, does not conform to the parties’ prior common intention.

 

The court held (at para 346) that Olson “established the necessary elements to a standard of ‘convincing proof’ for rectification of the contract in a case of mutual mistake.” More specifically, the court held (at para 336) that the parties had a common intention that “the Agreed Fence Height Restriction be the fencing provision due to Olson’s requirements for containing bison on the Property”; that this common intention was unchanged at the date the agreement was executed; and that the challenged document, that is, the Conservation Easement Amending Agreement “contains a mistake and does not conform to the parties’ intentions and prior oral agreement that the applicable fencing provision be the Agreed Fence Height Restriction” (paras 338-339).

 

The subsequent conduct of the parties corroborates and supports the conclusion that all three elements required to rectify a contract on the basis of mutual mistake are established in this case. Such conduct includes the fact that subsequent leases did not contain fence height restrictions, as well as the fact that, soon after purchasing the property and prior to discovering the mistake in integration of the oral agreement, Olson began the construction of a new fence “as permitted by the Agreed Fence Height Restriction”, that is, the oral agreement (at para 342).

 

Moreover, the parties had a common understanding of Olson’s purpose for acquiring the ranch (para 345). As the court observed (at para 340), Olson would never have agreed to the inclusion of the fence height restriction in the Conservation Easement Amending Agreement because “it would have precluded their running wild bison on the Property.” Thus the fence height restriction urged by the NCC is not only “contrary to the evidence”, but it also “defies logic for a wild bison ranch”.

 

Ultimately, the court granted an order rectifying the Conservation Easement Amending Agreement on the basis of mutual mistake.

 

Unilateral Mistake

 

In the alternative, the court considered whether the agreement should be rectified on the basis of unilateral mistake. A unilateral mistake occurs where one party is mistaken as to a fundamental matter underlying a contract and the non-mistaken party is aware of the mistake made by the mistaken party. (See, for example, Glasner v. Royal Lepage Real Estate Services Ltd., 1992 CanLII 975 (BCSC)).

Numerous “high hurdles” stand in the way of a party seeking rectification of a contract on the basis of unilateral mistake (Performance Industries at paras 35-41). A contract induced by unilateral mistake may be rectified by the court if the party seeking rectification proves:

1. the existence of a prior oral agreement with definite and ascertainable terms;

2. that the written document does not correspond with the prior oral agreement (that is, that the terms agreed to orally were not written down properly);

3. that at the time of execution of the contract the defendant either knew or ought to have known of the mistake in reducing the oral terms to writing and the plaintiff did not. That is, that the defendant’s attempt to rely on the erroneous written document amounts to “fraud or the equivalent of fraud”;

4. their case on the basis of ‘convincing proof’

(Performance Industries at paras 31, 35-41).

Flowing from these ‘high hurdles’, a party seeking rectification of a contract on the basis of unilateral mistake bears the obligation to show the ‘precise form’ in which the written agreement can be made to conform to the parties’ prior oral agreement (Performance Industries at para 40).

Applying these principles to the facts, the court held (at para 349) that Olson convincingly established “the existence and content of a prior oral agreement between the parties on the Agreed Fence Height Restriction, which is inconsistent with the Replacement Fence Height Restriction actually contained in the written agreement”. Furthermore, the court held (at para 354) that the NCC either knew or ought to have known of the mistake in reducing the oral terms to writing. This is so because the NCC knew that Olson’s intended uses for the Property “depended on his ability to contain wild bison within its boundaries” (para 354). Thus, had the NCC reviewed the Conservation Easement Amending Agreement they “would have recognized the mistake in it” (para 354). Finally, the court held (at para 355) that, the NCC’s attempt to rely on the erroneous written document amounts to ‘fraud or the equivalent of fraud’. Thus, while this transaction fell short of deceit, “it would be unconscionable for the NCC to avail itself of the advantage obtained” through the erroneous Conservation Easement Amending Agreement. In the end, the court held that the Conservation Easement Amending Agreement should be rectified on the basis of unilateral mistake.

Commentary

The order of rectification granted by the court would appear generous but it is supportable on the facts. There is no doubt that the parties agreed orally to amend or modify the fence height restriction in the initial conservation easement agreement. This conclusion is supported by the subsequent conduct of the parties, particularly the post-sale grazing lease which did not contain fence height restrictions. Moreover, Olson would not have purchased the property if the NCC had insisted on the terms of the initial conservation easement agreement as the basis for the sale and purchase of the property.

The problem, however, is that the court’s description of the precise terms of the oral agreement appears to be more elaborate than the terms urged by Olson at trial. Olson argued in his closing brief that the parties agreed orally on fencing as follows:

[124] The Grantor may maintain, replace and repair the fences, roads, buildings and other improvements located on the Property as of the date of this easement. The fences and roads are to be maintained, replaced or repaired.

Yet the court found the oral agreement to be more elaborate than Olson urged in his pleading (paras 126 & 327). By making a determination which apparently goes beyond Olson’s specific pleading, the court opens itself to the criticism that it drafted the terms of the ‘Agreed Fence Height Restriction’ for the parties. It also makes the court susceptible to a charge of overreaching its jurisdiction. As the Supreme Court of Canada noted in Performance Industries (at para 40), the court’s equitable jurisdiction to rectify a contract “is limited to putting into words that – and only that – which the parties had already orally agreed to.”

It should also be said that the court gave very little consideration to the issue of negligence in this case. In fact, the court alluded fleetingly to the issue of negligence by observing (at para 361) that “Rectification here is not a belated substitute for due diligence; it would not unjustly impose liability on the NCC that ought more properly to be attributed to Olson’s negligence.” Although “due diligence on the part of the plaintiff is not a condition precedent to rectification”, negligence is a factor to be considered by the court in determining whether to rectify a contract (Performance Industries at para 66). In the instant case, the court ought to have considered whether Olson’s negligence was of such a nature as to deny him the equitable remedy of rectification.

That being said, Olson’s negligence appears to be extenuated by several factors including the peculiar circumstances surrounding his execution of the purchase agreement. First, Olson’s mistaken assumption was directly attributable to his solicitor to whom he had given clear instructions to ensure that the written agreement accords with the prior oral agreement he reached with the NCC. Second, Olson signed the purchase documents in hurried circumstances primarily because the NCC wanted the documents “right away to convenience its internal needs” (para 121). As the court found,

[121] On August 28, 2003, Olson was departing with his family on vacation, but delayed his departure to get his offer in on the Property to accommodate the NCC. The NCC wanted it right away to convenience its internal needs. Olson came into his office to sign the [Offer to Purchase], with his large family waiting outside in their vehicle. Although the [Offer to Purchase] did not yet accurately reflect the agreement Olson reached with Green on the various remaining amendments they negotiated to the Initial ACA CE, he entrusted the remaining corrections to be made by passing on his comments to his commercial lawyer, …. Olson signed the [Offer to Purchase] in blank and rejoined his family.

In these circumstances, it would be harsh to deny Olson the equitable remedy of rectification simply because Olson and his solicitor were negligent. Moreover, as the court observed (at para 361), despite Olson’s negligence, an order of rectification “would not unjustly impose liability on the NCC”.

Finally, it should be observed that the court also dealt with the question of whether the title to the Penny Ranch should be rectified pursuant to section 190(1) of the Land Titles Act (paras 405-431). However, this aspect of the decision is not the focus of this post.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

 

Punitive Damages Now Possible in Alberta in Fatal Accident Actions

Thu, 08/07/2014 - 10:00am

By: Iwan Saunders

PDF Version: Punitive Damages Now Possible in Alberta in Fatal Accident Actions

Case commented on: Steinkrauss v Afridi, 2013 ABCA 417, as clarified at 2014 ABCA 14

As a result of Steinkrauss v Afridi in the Court of Appeal, punitive damages are now possible in Alberta in fatal accident actions.  This post looks at three things: the background to Steinkrauss,what the case means for this and future claimants, and why the Alberta Legislature should fall in line with Steinkrauss and change the law regarding survival actions.

Background to Fatal Accident Actions and Claims for Punitive Damages

At common law survivors had no right of action whatsoever for their own losses through another’s wrongful death, a rule originally established in England in Baker v Bolton in 1808, 170 ER 1033 (KB), where a husband failed to recover anything for the death of his wife in a stagecoach accident.  Eventually the rule was reformed, by a statute colloquially known after its sponsor as Lord Campbell’s Act: An Act for Compensating the Families of Persons Killed by Accidents, 1846, 9 & 10 Vict, c 93.  This Act was immediately imported by the then province of Canada, 10 & 11 Vict, c 6 (1847), and now, in one form or another, all Canadian provinces and territories have similar legislation of their own.  [For analysis of this legislation and of fatal accident actions generally, see my chapters in Ken Cooper-Stephenson, Personal Injury Damages in Canada (2d edition, Carswell 1996), chapters 10 and 11 (631-49, and 651-720).]

In Alberta, by virtue of the Fatal Accidents Act, RSA 2000, c. F-8 (as amended), an action is now possible on behalf of survivors (sometimes called beneficiaries or dependents) where, to quote from section 2, “the death of a person has been caused by a wrongful act, neglect or default”.  But the right to sue is limited.  First, by virtue of other language in section 2, survivors can only sue if the deceased herself could have sued, and sued successfully, were it not for her death; to this extent the survivors’ action is derivative.  Secondly, only authorized survivors can sue (see in particular sections 3(1) and 8(1)), and thirdly, they can only sue for authorized damages.  The basic damages section is section 3(1), which provides that “the court may give to the persons respectively for whose benefit the action has been brought those damages that the court considers appropriate to the injury resulting from the death”, though, in addition, section 7 allows recovery of some specified “expenses and fees”, while section 8 provides fixed sums for “grief and loss of the guidance, care and companionship of the deceased person”.

With respect to punitive damages, sometimes called exemplary damages, they are of course quite different from compensatory damages.  Whereas compensatory damages focus on plaintiffs and their actual and potential losses, punitive damages focus on defendants and their extremely bad behavior. They are tantamount to a civil fine, their goal being to punish, deter and denounce wrongdoers rather than to compensate victims.  At common law, in England they go back to 1763 (Huckle v Money, 2 Wils KB 205, 95 ER 768 (1763); Wilkes v Wood, Lofft 1, 98 ER 489 (KB 1763)), while in Canada the Supreme Court has regularly approved of them, the dominant authority today being Whiten v Pilot Insurance Co., 2002 SCC 18.

As for punitive damages in fatal accident actions, in New Brunswick they are expressly recoverable “in appropriate cases”, though, if awarded, “they shall be for the benefit of the deceased’s estate”: Fatal Accidents Act, SNB 2012, c 104, s 17.  Elsewhere in Canada, including Alberta, none of the applicable statutes explicitly mentions them, though they have been decisively rejected by courts in British Columbia, Ontario and Nova Scotia: Allan Estate (Executors of) v. Co-Operators Life Insurance Co., 1999 BCCA 35; Campbell v Read, 1987 BCCA 2402; Lord v Downer, 1999 ONCA 1875; Latimer v Canadian National Ry. Co., 2007 ONSC 5689; Rowe v Brown, 2008 NSSC 13. On the other hand, two trial judgments in Saskatchewan and Prince Edward Island contain inconclusive dicta arguably approving them: Anderson v Board of Education (1986), 50 Sask R 4 at 5 (QB), and Blacquiere’s Estate v Canadian Motor Sales Corp. (1975), 10 Nfld & PEI R 178 at 210 (PEI SC).

This brings us to Steinkrauss v Afridi.

Facts and Decisions in Steinkrauss

Deirdre Steinkrauss was a cancer patient who sued her doctor for medical malpractice, alleging that he had negligently failed to diagnose her cancer as early as he should have, having omitted to carry out certain genetic and other testing.  When Deirdre died shortly after commencing her action, her husband sought to prosecute it in his own name, on behalf of himself and their children, essentially under the Fatal Accidents Act (the FAA).  Among proposed amendments to the original Statement of Claim, he included a prayer for punitive damages, asserting that, sometime after Deirdre’s diagnosis but before her death, the defendant tried to shift responsibility from himself to Deirdre by altering her medical charts to indicate that he had suggested the testing to her and that she had declined. A Master in Chambers found for the plaintiff and allowed his amendments.  But on appeal a chambers judge (Gates, J.) struck out the prayer for punitive damages, holding that the language of the FAA did not permit them: Steinkrauss v Afridi, 2013 ABQB 179.  The plaintiff then appealed to the Court of Appeal and won.  In a Memorandum of Judgment (2013 ABCA 417, written by Berger, Slatter and Veldhuis, JJA), the Court decided that punitive damages are indeed possible under the FAA and that in this case, on the facts pled, the plaintiff was duly entitled to pursue his claim at trial.

Comment

The plaintiff in Steinkrauss can clearly sue under the FAA for compensatory damages, assuming of course he proves that the defendant negligently caused the deceased’s death, or perhaps more accurately her premature death.  Assuming the defendant’s negligence, the deceased herself could have sued for her own losses, had she not died, and so now the plaintiff can sue under the statute for his and his children’s losses.  That said, the ruling that he could also sue for punitive damages came as something of a surprise, at least to me.

As mentioned, the governing language in the FAA is in section 3(1), which authorizes judges to give “those damages that the court considers appropriate to the injury resulting from the death” (emphasis mine).  While this is evidently poor drafting, I had always taken it to mean that fatal accident actions were exclusively about compensating dependent loss and not at all about punishing, deterring or denouncing defendant misbehavior; in other words, that the statutory provision as given could never sustain punitive awards.  And beyond me, the ruling may well have surprised some personal injury lawyers too.  In one case for example, plaintiff counsel felt obliged to concede, even during an interlocutory application, that punitive claims under the FAA were simply impossible: Clapperton Estate v Davey, 2009 ABQB 63, at para 2.  But obviously those of us who thought this way were mistaken all along.  For the Court of Appeal in Steinkrauss has plainly held that section 3(1) is in fact “open-ended” (2013 ABCA 417 at para 12), that punitive damages are indeed allowable, and that in every case the basic question is whether they are “appropriate to the injury resulting from the death” (para 19).  “While it is arguable this test will be difficult to meet,” said the Court, “there is no reason to foreclose recovery in every case as a matter of law” (again, para 19).

Leaving aside the dispute over statutory wording and legislative intention, as a matter of policy Steinkrauss makes sense.  The argument is neither difficult nor new. [See originally Ken Cooper-Stephenson and Iwan Saunders, Personal Injury Damages in Canada (Carswell 1981) at 640].  If punitive damages are socially useful in common law actions, as the Supreme Court consistently says they are, as a supplement to the criminal law, prima facie they are socially useful in statutory actions too.  Given misconduct deserving a penalty, the cause of action per se ought not to matter. Thus in the context of Steinkrauss, if fines can be levied inter vivos for wrongfully injuring people, they should also be levied via the statute for wrongfully killing them.  This is what the Court of Appeal is presumably saying.

Anyway, granted the reality of punitive claims under the FAA, the question for judges now is how best to regulate them.  The statute being essentially silent on this, the natural place to get some guidance is the common law, particularly Whiten v Pilot Insurance (above) and its extensive progeny.  Basically the key starting points at common law are these.  First of all defendants are never subject to civil fines simply for committing civil wrongs.  Punitive damages are only awarded in exceptional cases, where the defendant’s conduct was sufficiently egregious (flagrant or blameworthy) to warrant a fine: see Whiten, cited in Steinkrauss (2013 ABCA 417 at paras 14 and 20).  Furthermore, as Lord Devlin wisely cautioned in the famous case of Rookes v Barnard, the plaintiff can only recover punitive damages if he is “the victim of the punishable behavior”. Otherwise, “[t]he anomaly inherent in exemplary damages would become an absurdity if the plaintiff totally unaffected by some oppressive conduct … obtained a windfall in consequence”: [1964] AC 1129 at 1227 (HL).  Or as Lord Diplock put it rather more vividly in a later case, the plaintiff “can only profit from the windfall if the wind was blowing his way”: Cassell & Co. Ltd. v Broome, [1972] AC 1027 at 1126 (HL).  According to the cases, what this latter point means in practice is this.  Plaintiffs must show that defendants violated their rights, by egregiously committing an actionable wrong against them, whether in tort, contract or equity.  On the other hand plaintiffs need not show (a) that defendants aimed their conduct specifically at them, nor (b) in the case of wrongs actionable per se, that they suffered actual loss.  So long as there was egregious misconduct in the course of committing a wrong to the plaintiff, punitive damages can follow.

In Steinkrauss the Court gave some direction on judging punitive claims under the FAA and in doing so drew on the common law, though not always in so many words. The Court explicitly cited Whiten (2013 ABCA 417 at paras 14 and 20), particularly on the matter of egregious misconduct, and also, as I read the judgment, implicitly required that plaintiffs be victims of the punishable behaviour.  According to the Court, whether claimants can get punitive damages under the FAA “depends on the source of the claim”, and in that regard “the key question will be whether [the] egregious conduct was sufficiently connected to the claim of the dependants arising from the death, such that it can be said to be ‘appropriate to the injury resulting from the death’” (again, para 14).  And then some months later, on an application for clarification of these dicta, the Court essentially affirmed what it had originally said: see 2014 ABCA 14 at para 4.  As the Court saw it, the plaintiff’s position at trial would come down to this: the defendant’s conduct in falsifying the records had impeded his ability to prove the defendant’s negligence, thus making the conduct sufficiently connected to his own action to justify his claim for an exemplary sum.  That position, held the Court, was at least arguable, even if the fraud occurred sometime after the negligence and before the deceased’s death, rather than before the negligence or after the death.  Even on that version of the facts, the plaintiff was still entitled to his day in court.

If Steinkrauss does go to trial and the plaintiff does prove evidence tampering as alleged, this could help him prove some of the facts necessary for negligence. Besides its effect on the defendant’s general credibility, the tampering may qualify as spoliation, thus giving rise to a rebuttable presumption of fact that the original medical records would have told against the defendant: on spoliation, see McDougall v Black & Decker Canada Inc., 2008 ABCA 353.  And if the plaintiff goes on to succeed in his action, by duly establishing negligence, he will presumably ask for solicitor-client costs as an indirect sanction for the tampering.  But whether the plaintiff can get punitive damages for that tampering seems questionable, even if the tampering is considered egregious.

The reason seems fundamental, whether the action is at common law or, as here, via a statute.  Civil fines should only be levied for civil wrongdoing and nothing less, in much the same way that criminal fines can only be levied for criminal wrongdoing and nothing less. However reprehensible the defendant’s misconduct, he should only be fined for it in a civil court if it formed part of the actionable wrong on which the litigation is founded.  [See for example Eli Lilly and Company v Apotex Inc., 2009 FC 991 at paras 657-64, citing Whiten.]  In Steinkrauss, while his wife’s death, if wrongful, was also a wrong to the plaintiff, there was no flagrant misconduct surrounding the negligence, so far as we know.  As for the evidence tampering, it apparently occurred after the negligence and was quite distinct from it, therefore falling outside the wrong on which this plaintiff necessarily bases his action.  Of course, a trial judge under the FAA might ignore the common law and give punitive damages anyway, simply on the ground that the tampering was egregious and, to quote from the Court of Appeal, “sufficiently connected to the claim of the dependants arising from the death, such that that it can be said to be ‘appropriate to the injury resulting from the death’”.  But if so, is this what the Court of Appeal really had in mind?

Subject to his pleadings and limitation periods, the plaintiff could alternatively assert the tampering (and punitive damages) in a separate cause of action.  But what separate cause of action does he have?  The tort of deceit (fraudulent misrepresentation) comes to mind but seems highly unlikely here, since deceit traditionally requires actual plaintiff reliance; a mere attempt to deceive the plaintiff is not enough.  One possibility is a new tort of spoliation of evidence.  Some American courts have already adopted it and some Canadian courts already seem at least open to it: see for example, McDougall, above; Holland v Marshall, 2008 BCCA 468; Spasic Estate v Imperial Tobacco Ltd., 2000 CanLII 17170 (ON CA).  But so far there is no such tort in Alberta or in any other Canadian jurisdiction for that matter.  So the plaintiff in Steinkrauss would have to pioneer it.  [The British Columbia Law Institute has proposed such a tort: Report on Spoliation of Evidence (BCLI Report No. 34, 2004).]

Punitive Damages in Survival Actions

As a result of Steinkrauss, there is now an inconsistency between the FAA and its relative, the Survival of Actions Act, RSA 2000, c S-27 (as amended) (the SAA). The solution lies in amending the SAA.

By virtue of legislation in all provinces and territories, reversing the common law, civil actions now survive the death of plaintiffs for the benefit of their estates.  [For analysis of this legislation and of survival actions generally, see my chapter in Ken Cooper-Stephenson, Personal Injury Damages in Canada (2d edition, Carswell 1996), c. 12 (721-46).]  And in principle, I suggest, claims for punitive damages should survive too.  Once again the argument is familiar.  Punitive damages are about the defendant’s conduct, not the plaintiff’s injury and loss. If the defendant’s misconduct while committing the wrong was sufficiently egregious to warrant a fine inter vivos, the plaintiff’s death should make no difference and the fine should still be imposed. While admittedly punitive damages would be a windfall to the estate, they are something of a windfall to all successful plaintiffs. Nevertheless we allow them because they are socially useful and plaintiffs perform a socially useful service by claiming them: see Binnie J in Whiten, above, at para 37.

However in Alberta the law is otherwise. Section 5(2)(a) of the SAA explicitly precludes “punitive or exemplary damages” in all circumstances.  So in Steinkrauss, while the deceased’s survivors could claim punitive damages under the FAA, the deceased’s estate could not have done so under the SAA. Thus the defendant relied on section 5(2)(a) of the SAA as a reason for prohibiting awards under the FAA, an argument rightly rejected by the Court of Appeal.

As I see it, the Alberta Legislature should now amend the SAA to allow punitive claims, in actions both by and against estates.

[For a Quebec case in the Supreme Court approving punitive damages against a deceased wrongdoer’s estate, see de Montigny v Brossard (Succession), 2010 SCC 51. And for a comment supporting the decision in Brossard, see Nicholas Rafferty and Iwan Saunders, “Developments in Contract and Tort Law: The 2010-11 Term” (2011), 55 Sup Ct L Rev (2d) 163 at 196-202.]

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

 

Grassy Narrows, Division of Powers and International Law

Wed, 08/06/2014 - 10:35am

By: Nigel Bankes

PDF Version: Grassy Narrows, Division of Powers and International Law

Case commented on: Grassy Narrows First Nation v Ontario (Natural Resources), 2014 SCC 48

This post discusses two issues arising from the Supreme Court’s decision in Grassy Narrows. The post first considers the implications of the Court’s conclusion that the doctrine of interjurisdictional immunity does not apply in a case where a province infringes the treaty right to hunt leaving the treaty party with no meaningful right to hunt. Second the post argues that the Court’s conclusion that a provincial government may be able to justify an infringement of hunting rights of this nature is inconsistent with Canada’s obligations under international law.

The division of powers issue               

Prior to this decision and the Court’s decision in Tsilhqot’in Nation v British Columbia, 2014 SCC 44 I think that it was broadly understood that a province could not impair an Indian treaty right. A province could not do so directly because such a law would be a law in relation to Indians or lands reserved for Indians. Perhaps the best discussion of this issue is Justice Davey’s decision in the British Columbia Court of Appeal in R v White and Bob (1964), 50 DLR (2d) 613 (BCCA), aff’d [1965] SCR vi, which dealt with a charge of being in possession of game out of season and without a permit under the BC Wildlife Act. The accused defended the charge principally on the basis of one of the Douglas Treaties to which their ancestors were a party. The Douglas Treaties contained the following clause:

The condition of, or understanding of this sale, is this, that our village sites and enclosed fields, are to be kept for our own use, for the use of our children, and for those who may follow after us, and the lands shall be properly surveyed hereafter; it is understood however, that the land itself with these small exceptions, becomes the entire property of the white people forever, it is also understood that we are at liberty to hunt over the unoccupied lands, and to carry on our fisheries as formerly.

While it is arguable that Justice Davey’s judgement principally turns on the opening language of what is now s.88 of the Indian Act, RSC 1985, c I-5, he also went on to state as follows:

Sections 8 … and 15 … of the Game Act specifically exempt Indians from the operation of certain provisions of the Act, and from that I think it clear that the other provisions are intended to be of general application and to include Indians. If these general sections are sufficiently clear to show an intention to abrogate or qualify the contractual rights of hunting notoriously reserved to Indians by agreements such [this Treaty] they would, in my opinion, fail in that purpose because that would be legislation in relation to Indians that falls within Parliament’s exclusive legislative authority under s. 91 (24) of the B.N.A. Act, and also because that would conflict with s. 87 [now s.88] of the Indian Act passed under that authority. Legislation that abrogates or abridges the hunting rights reserved to Indians under the treaties and agreements by which they sold their ancient territories to the Crown and to the Hudson’s Bay Company for white settlement is, in my respectful opinion, legislation in relation to Indians because it deals with rights peculiar to them. Lord Watson’s judgment in St. Catherine’s Milling & Lumber Co. v. The Queen (1888), 58 L.J.P.C. 54, if any authority is needed, makes that clear. At p. 60 he observed that the plain policy of the B.N.A. Act is to vest legislative control over Indian affairs generally in one central authority. On the same page he spoke of Parliament’s exclusive power to regulate the Indians’ privilege of hunting and fishing. In my opinion, their peculiar rights of hunting and fishing over their ancient hunting grounds arising under agreements by which they collectively sold their ancient lands are Indian affairs over which Parliament has exclusive legislative authority, and only Parliament can derogate from those rights.

There was also room for thinking that the province could not impair an Indian treaty right even indirectly; such a law would be inapplicable because of the doctrine of interjurisdictional immunity. The premise of the argument is that treaty rights are part of the core content of one or other head of s.91(24) and that any non-trivial interference with a treaty right would make a valid law not invalid but inapplicable. Authority for this proposition prior to Tsilhqot’in would have included R. v Morris, [2006] 2 SCR 915.

But in Tsilhqot’in (handed down the week before Grassy Narrows), the Supreme Court of Canada in an obiter statement apparently made the doctrine of interjurisdictional immunity inapplicable to head 24 of section 91 of the Constitution Act, 1867 at least in relation to aboriginal title (at para 140 et seq). It also expressed doubt about the authority of Morris on interjurisdictional immunity (at para 150). My colleague Jennifer Koshan and I have criticized that aspect of Tsilhqot’in in an earlier post here. In Grassy Narrows (in another obiter statement at para 53) the Court, again unanimously, extended the non-applicability of the doctrine of interjursidictional immunity to provincial laws and actions that infringe a treaty right. The Court did so with the bald statement (at para 33) that “Tsilhqot’in Nation v British Columbia is a full answer” to any claim as to the inapplicability of provincial laws. Thus, there is no longer a protected core to head 24 of s.91. Provincial laws and provincial alienations that are so extensive as to render an Indian treaty right (i.e. a right arising from an agreement, i.e. consent) to hunt meaningless will not be inapplicable. That however is not the end of the matter; while there is no bright line inapplicability rule the provincial government will still have to justify takings up of land that have this result.

The relevance of international law

This brings us to the second topic of this post, the relevance of international law. The argument I want to present here is that if the relevant government has used (taken up) or authorized the use of lands within the traditional territory of a First Nation to such an extent that a treaty right to hunt is no longer meaningful then, as a matter of international law, there is no opportunity to justify the infringement because the government action amounts to an impermissible denial of a minority’s right to culture within the meaning of Article 27 of the International Covenant on Civil and Political Rights (ICCPR). I have hinted at various forms of this argument in previous posts (here, here and here) but I will now present a more elaborate form of the argument.

Article 27 provides as follows.

Article 27

In those States in which ethnic, religious or linguistic minorities exist, persons belonging to such minorities shall not be denied the right, in community with the other members of their group, to enjoy their own culture, to profess and practise their own religion, or to use their own language.

While the Article does not refer specifically to indigenous peoples it is clear that in most if not all cases indigenous peoples will qualify as minorities within the meaning of the Article. The Article protects the minority’s right to culture. In its interpretive note on Article 27 (General Comment No. 23) the Human Rights Committee (HRC) (the expert supervisory body for the ICCPR) has emphasized that in the case of indigenous communities the right to culture may have a material element that is closely connected to traditional territories:

7. With regard to the exercise of the cultural rights protected under article 27, the Committee observes that culture manifests itself in many forms, including a particular way of life associated with the use of land resources, especially in the case of indigenous peoples. That right may include such traditional activities as fishing or hunting and the right to live in reserves protected by law. The enjoyment of those rights may require positive legal measures of protection and measures to ensure the effective participation of members of minority communities in decisions which affect them. (references omitted)

In its decisions the HRC has also emphasized that not every state authorized activity in a community’s traditional territory will violate Article 27. However, the HRC does contemplate that the cumulative effect of state authorized activities may amount to a denial of the right to culture. For example, in a case involving water transfers in Peru that affected the traditional activities of an indigenous community (Angela Poma Poma v Peru (2009)), the HRC discussed the threshold question as follows (at paras 7.2 – 7.7):

… the exercise of the cultural rights protected under article 27 … manifests itself in many forms, including a particular way of life associated with the use of land resources, especially in the case of indigenous peoples.  That right may include such traditional activities as fishing or hunting and the right to live in reserves protected by law.  The enjoyment of those rights may require positive legal measures of protection and measures to ensure the effective participation of members of minority communities in decisions which affect them.  The protection of these rights is directed to ensure the survival and continued development of cultural identity, thus enriching the fabric of society as a whole.

The Committee recognizes that a State may legitimately take steps to promote its economic development.  Nevertheless, it recalls that economic development may not undermine the rights protected by article 27.  Thus the leeway the State has in this area should be commensurate with the obligations it must assume under article 27.  [M]easures whose impact amounts to a denial of the right of a community to enjoy its own culture are incompatible with article 27, whereas measures with only a limited impact on the way of life and livelihood of persons belonging to that community would not necessarily amount to a denial of the rights under article 2.

…  The Committee also observes that the author has been unable to continue benefiting from her traditional economic activity owing to the drying out of the land and loss of her livestock.  The Committee therefore considers that the State’s action has substantively compromised the way of life and culture of the author, as a member of her community.  The Committee concludes that the activities carried out by the State party violate the right of the author to enjoy her own culture together with the other members of her group, in accordance with article 27 of the Covenant.

Thus, if an indigenous community can show that the right to hunt in traditional territory (perhaps following a season round, see Dick v R, [1985] 2 SCR 309) is core to their right to culture; and if the state (provincial or federal government) has authorized activities within a community’s traditional territory that are so extensive that the community no longer has a meaningful right to hunt and has therefore been denied its right to culture, then the state has no option but to curtail those activities that interfere with the right to culture. In other words, in this case international law supports a bright line rule and would not support the argument that a provincial government should have the opportunity to further justify its infringement of a treaty right (or indeed an aboriginal harvesting right) where state licensed activities rendered the right to hunt meaningless. In sum, the approach sanctioned by the Supreme Court of Canada in Grassy Narrows is inconsistent with Canada’s international obligations under Article 27 of ICCPR. Another and softer way to put the argument is that when a province seeks to justify an infringement of the right to hunt, the indigenous community should plead that Canada’s obligations under international law should inform that justificatory exercise: Baker v Canada, [1999] 2 SCR 917.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

 

 

Canada and Nova Scotia Finalize Equivalency Agreement on the Control of Greenhouse Gas Emissions in the Electricity Sector

Tue, 08/05/2014 - 10:00am

By: Nigel Bankes

PDF Version: Canada and Nova Scotia Finalize Equivalency Agreement on the Control of Greenhouse Gas Emissions in the Electricity Sector

Documents commented on:  Agreement on the Equivalency of Federal and Nova Scotia Regulations for the Control of Greenhouse Gas Emissions from Electricity Producers in Nova Scotia;Proposed Order in Council Declaring that the Reduction of Carbon Dioxide Emissions from Coal-fired Generation of Electricity Regulations do not apply in Nova Scotia, Canada Gazette vol 148 (2014), June 28, 2014 and the accompanying Regulatory Impact Analysis Statement

This Agreement is the first greenhouse gas (GHG) equivalency agreement to be finalized between Canada and a province. The Agreement and the accompanying draft Order in Council will serve to suspend the application of Canada’s Reduction of Carbon Dioxide Emissions from Coal-Fired Generation of Electricity Regulations, SOR/2012-167 (the federal Coal Regulations or the CFGRs) made under s.93 of the Canadian Environmental Protection Act, 1999, SC 1999, c 33 (CEPA, 1999) in the province of Nova Scotia. For comment on the CFGRs see the post by Astrid Kalkbrenner here. The Agreement will be of considerable interest to other jurisdictions (including Alberta) which are negotiating equivalency agreements with Canada to avoid the application of federal GHG regulations. While a draft of the Agreement has been available for a couple of years (see here, and for a very short summary of the two supportive comments received see here), and there are very few changes between the draft and the final version, what is new is the release of the Regulatory Impact Analysis Statement (RIAS) which casts some light on the methodology that the federal government will apply in assessing equivalency.

Interested parties have 60 days to provide comments and/or file a notice of objection requesting establishment of a board of review under s.333 of CEPA, 1999.

Federal Greenhouse Gas Regulations

As is well known, the federal government proposes to reach its Copenhagen commitment to reduce greenhouse gas emissions by implementing a sector by sector regulatory approach. This has proven to be painfully slow. The federal government adopted the CFGRs in 2012 but promised regulations for other sectors, including the oil and gas sector, have been repeatedly delayed. The CFGRs establish a performance standard of 420 tonnes of carbon dioxide per gigawatt-hour (tCO2/GWh) for new coal-fired electricity generation units and those that have reached the end of their useful life. According to the RIAS, the application of the CFGRs would require the province (Nova Scotia Power) to retire its coal fired units totalling 952 MW of capacity between 2020 and 2030. Nova Scotia sought to avoid this result (with the implication that it would need to build new natural gas capacity) and therefore entered into negotiations for an equivalency agreement.

Equivalency

The CFGRs are enacted pursuant to the Canadian Environmental Protection Act, 1999. Section 10 of CEPA, 1999 allows the Governor in Council to make an order declaring that specific regulations “do not apply” to a particular jurisdiction where the Minister of Environment and the government of the other jurisdiction “agree in writing” that there are laws in force in the other jurisdiction, the provisions of which “are equivalent to” the regulations. The premise of the Nova Scotia Agreement is that an “outcome-based” equivalency agreement is adequate to meet the terms of s.10 thereby justifying an order making the federal regulations inapplicable. Thus, the purpose of the Nova Scotia Agreement is to allow existing coal fired facilities to continue generating even though they will not be able to meet the standards prescribed by the federal regulations. A finding of equivalency is justified on the basis that Nova Scotia will achieve reductions equivalent to those that will be achieved through the federal regulations by reduced generation at coal facilities and by other measures that have nothing to do with the coal sector. These other measures include a commitment to increase the share of renewable energy in the province’s energy mix through a number of initiatives including a feed in tariff program, and support for the Muskrat Falls hydro project, as well as various energy efficiency programs including smart metering. In addition, Nova Scotia was also required to amend its regulations to ensure that electricity producers must meet certain caps prescribed in the Agreement. This is perhaps the crucial element here – the adoption of a hard cap for the sector rather than the technology based target – because it provides some certainty as to reduced emissions in the sector.

If we assume that the outcomes based approach to equivalency is appropriate (and it is not entirely clear that it is, since s.10 of CEPA 1999 focuses on the provisions of the laws of the other jurisdiction, not on the effect or outcome of the measures taken in the other jurisdiction) it is of course necessary to show that the provincial measures will reduce GHG emissions by at least as much as the reductions that would be achieved by application of the CFGRs. It is evident that this assessment requires the construction of two scenarios. The first is a business as usual (BAU) scenario in which emissions are assessed on the basis that the equivalency agreement is not concluded and thus that the CFGRs will apply as well as any applicable provincial rules (other than incremental commitments made as part of the equivalency agreement – see s.8.2.3 of the RIAS and the “caps” that the province undertook to impose). The second scenario, the regulatory scenario, establishes what the electricity generation sector is expected to look like with the implementation of the proposed Order in Council (i.e. the federal regulations are backed out and the incremental provincial commitments apply).

The RIAS summarizes the comparison between the two scenarios (here I focus on the reduced emission rather than the more general cost-benefit analysis, which, for example, assumes avoided decommissioning costs at coal generation facilities by pushing them out beyond the period of review, 2021 – 2030) as follows (RIAS, s.8.2.3):

Standing down the federal coal-fired electricity regulations would allow the electricity producer to continue to operate coal-fired units reaching the end of their useful life that would otherwise have been closed in the BAU scenario at the end of their useful life. This flexibility is expected to result in avoided natural gas capacity investments. ……

Also, prior to 2030, the electricity producer is expected to generate, relative to the BAU scenario, less electricity from coal-fired units and natural gas units in order to comply with provincial GHG emission caps for 2021–2030. Therefore, the electricity sector is expected to show a cumulative reduction in generation in Nova Scotia and correspondingly in Canada. It is also expected that Nova Scotia would need to rely more on importing renewable energy from the Muskrat Falls project in order to meet demand and the provincial renewable energy standards. This reduces, relative to the BAU scenario, the generation available for export to the United States from Muskrat Falls through New Brunswick, leading to an overall reduction in electricity exports. The reduction in generation from coal-fired units and natural gas units also leads to additional benefits from GHG emissions reduction, additional changes in air pollutant emissions, avoided fuel costs, avoided variable O&M costs associated with less reliance on coal-fired units and natural gas units.

The RIAS also provides an analysis of changes in demand (RIAS s.8.4.1) (a small reduction in demand in the regulatory scenario presumably because of the emphasis on demand side management (DSM) measures), and changes in the generation mix to meet that demand (RIAS, s.8.4.3) (projects less investment in new natural gas capacity in the regulatory scenario). The overall analysis is that over the 2015–2020 period, the regulatory scenario is “expected to result in an incremental reduction of 0.3 Mt CO2e of GHG emissions” and over the 2021–2030 period “incremental GHG reductions are estimated to be about 3 Mt CO2e of GHG emissions”.

Commentary

The principal difficulty with any scenario analysis such as this is the same difficulty that faces any offsets analysis and that is the challenge of showing additionality (especially over such a long time frame). For example, one might expect that a utility would take DSM measures in any event to avoid or defer building new capacity and so it is not clear why any reductions associated with DSM should be credited to the regulatory scenario. Similarly, the BAU analysis assumes that natural gas generation would be added to replace retired coal generation but at this stage we do not know what any sector specific regulations might look like for the natural gas sector. This might of course influence what is business as usual and what is therefore additional – that which is required by law is not additional.

I think that this points to the general difficulty of negotiating equivalency agreements on a sector by sector basis where the potential equivalent commitments that provinces may offer involve activities in other yet-to-be-regulated industrial sectors; it would be perverse if a province were offering equivalency commitments from activities undertaken in another sector which then falls to be covered by new federal regulations for that sector.

One paragraph was added to the final version of the Equivalency Agreement which goes some way to addressing this issue:

4.3 The Parties agree that, should they develop further regulations relating to greenhouse gases or air pollutants affecting the electricity sector in Nova Scotia, they will in good faith use their best efforts to conclude an equivalency agreement in respect of those regulations.  If the regulation pertains to greenhouse gases, the Parties may choose to amend the present agreement and determine equivalency on the basis of a comparison of the aggregate impact of both the Reduction of Carbon Dioxide from Coal-Fired Generation of Electricity Regulations and any new federal regulation on electricity sector emissions in Nova Scotia with the impact of the Greenhouse Gas Emissions Regulations on electricity sector emissions in Nova Scotia.

All of this of course is connected to the duration of the proposed arrangements and here the RIAS attempts to choreograph a merry dance between the competing needs of regime certainty (to obtain the needed investments and to avoid stranded assets) and flexibility (to respond to the sort of complexity referred to above as well as to provide the opportunity for adjustments should equivalency prove to be unattainable). The RIAS attempts to offer certainty by analysing emissions projections over a sixteen year period from 2015 – 2030 (the first year coinciding with the entry into force of the CFGRs and the Agreement and Order in Council and the second, more subjectively, based on “the time frame of the amendments made to Nova Scotia’s regulations to enter into the equivalency agreement for the control of GHG emissions from the electricity sector in Nova Scotia.” (RIAS, s.8.2.2)) The Agreement itself however tells a different story since it provides that it terminates on December 31, 2019 – or by either Party giving at least three months’ notice. The latter provision flies in the face of the need for certainty and that concern is only partially met by Canada’s commitment that if the Agreement is terminated before the end of 2019 Canada will “start applying” the CFGRs in the province “with as little disruption as possible in the circumstances, in a manner that takes into consideration the importance of a reliable supply of electricity that does not place an undue economic burden on Nova Scotia.” (s.5.4). That might be enough comfort in a province where the only coal fired generator is a Crown corporation; it might not be enough in a province like Alberta where all of the generation is privately owned. However, the challenge is built in to the Act; both the five year term and the three months notice to terminate are prescribed by s.10(8) of CEPA, 1999.

The Agreement addresses its possible extension or renewal as follows:

Should both Parties be in compliance with the Agreement on June 1, 2019, the Parties commit to initiate its renewal, provided that the effect on greenhouse gas emissions levels in the electricity sector in Nova Scotia of the limits to be imposed pursuant to the [provincial equivalent measures] is still considered equivalent to the effect on greenhouse gas emissions levels that would result from the application of the Reduction of Carbon Dioxide Emissions from Coal-Fired Generation of Electricity Regulations in Nova Scotia during this period.

It will be interesting to see if the finalization of this Agreement triggers the finalization of additional equivalency agreements. It will also be interesting to see if any party attempts to trigger the review procedure contemplated by s.10 and s.333 of the Act.

To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

Competition for Underground Disposal Space

Fri, 08/01/2014 - 10:00am

By: Nigel Bankes

PDF Version: Competition for Underground Disposal Space

Decision commented on: Alberta Energy Regulator, Canadian Natural Resources Limited, Application for Disposal, Lloydminster Field, 2014 ABAER 008

Previous posts on ABlawg have commented on sub-surface resource conflicts especially as between gas storage operations and oil and gas production: see Kallisto # 1 here and Kallisto # 2 here. This decision involving Canadian Natural Resources Limited (CNRL) illustrates that there may also be competition for good disposal sites and that the disposal operations of one party may affect the disposal operations of another party. While this decision involves the use of pore space for water disposal operations we can also anticipate competition for the use of pore space for carbon dioxide disposal purposes (i.e. carbon capture and storage, CCS) if that technology rolls out as anticipated to mitigate Alberta’s greenhouse gas emissions.

CNRL produces oil in the Lloydminster area which is associated with high volumes of produced water that it must dispose of. CNRL already has significant infrastructure in place for waste water transport and injection but its increased production in the area means that it also needs to expand its disposal capacity in order to avoid needing to shut in wells. Accordingly, CNRL identified a new disposal target in the form of the 3-17 well which was the subject of this contentious hearing and decision. CNRL first proposed to drill and complete this well for injection into the sandstone Dina formation but amended that application upon receipt of objections from Ener T (which owned the offsetting 16-7 injection well) so as to provide for completion in the Moberly and Cooking Lake Formations instead. The Alberta Energy Regulator (AER) licensed the well for these two formations but as a result of the poor injectivity that it encountered CNRL renewed its initial application asking for approval to abandon the 3-17 well in those two formations and recomplete it for injection purposes in the Dina Formation. Ener T renewed its objections resulting in the AER scheduling a hearing in which Ener T, having filed a statement of concern, was granted full participation rights.

The AER hearing panel identified four issues (at para 13):

    • the need for additional disposal capacity,
    • the evaluation of alternative disposal zones,
    • the capacity of the Dina Formation for injection fluids, and
    • the potential for adverse effects on Ener T’s AB/16-7 well.

On the question of need the Panel ruled as follows:

[20] The panel acknowledges that CNRL has productive oil wells that are shut in due to insufficient produced water disposal capacity. It accepts that CNRL’s ability to fully exploit and maximize the recovery of oil reserves from its land holdings in the area will be constrained if additional water disposal capacity is not obtained. The panel therefore finds there is sufficient evidence to support CNRL’s need for additional disposal capacity in the area.

On the question of alternatives the Panel noted that CNRL had explored a number of different options including the recompletion of abandoned wells but the Panel agreed with CNRL (at para 31) that “the existing wells reviewed may have wellbore integrity issues and that injecting into a new well, such as 3-17, will minimize risk with regards to containment.” The Panel also noted (at para 28) that CNRL’s proposal would tend to “minimize the effects on the landowner by eliminating the need for additional surface impacts of a new well or additional facilities.”

The Panel also examined the alternative formations that might be proposed for disposal. CNRL preferred the Dina formation because it was a proven disposal zone (with more than 300 wells in the region disposing into that zone). The Panel agreed with that assessment and noted (at para 40) that alternative formations proposed by Ener T would not typically be approved for disposal since they were potentially oil bearing.

The most interesting discussion in the decision relates to the potential for CNRL’s activities to harm Ener T’s injection operations at its offsetting 16-7 well. Here Ener T argued that CNRL’s operations would prejudice its own disposal activities and it therefore asked the AER to establish a protective set back or buffer zone to recognize its priority in point of time in identifying and developing this (disposal) resource:

[64] Ener T argued that it is not in the public interest to allow a large operator such as CNRL to effectively “wash out” a small operator’s operation. It also argued that it is not in the public interest to allow large operators like CNRL to select disposal locations that have been established by small operators to be effective disposal zones and drill its well in the same zone and vicinity as the small operator with no consideration of the effects. Instead, Ener T submitted that it is in the public interest to reward people that “get somewhere first” and “incur the risk and the cost of drilling into a certain area.”

To protect its interest Enert T asked the AER to impose a 1.6 km buffer on “competing” (my term) injection operations. In effect Ener T was seeking some sort of exclusive or at least prior entitlement to the disposal capacity of the Dina reservoir – which from the map included in Figure 1 of the decision would seem to go beyond the geography of its lease (although we actually learn nothing in this decision about Ener T’s production or its disposal rights).

The Panel rejected Enert T’s claim for relief. As for the specific request that the AER establish a buffer:

[74] The panel notes that the AER’s regulations do not restrict the distance between disposal wells. The AER’s Directive 065 does provide that notification of a disposal application is to be given to unit operators, approval holders, well licensees, mineral lessees and lessors, and landowners within a 1.6 km radius of a proposed disposal well where the disposal zone is known to be present.

More generally however it would seem that Ener T’s claim failed on the basis that it had not convinced the Panel of the likelihood of prejudice. The Panel thought that the evidence supported the idea that the target was “open and regionally extensive” rather than characterized by local barriers (at paras 62 and 76) and accordingly was likely a situation in which the reservoir should be able to accept disposal fluids from both wells. The Panel also thought that if there were any impacts on Ener T’s operations they would be limited because of the standard maximum wellhead injection pressure that would be imposed on CNRL’s operations. Given that the risk of adverse impact was low but with the positive benefits of increased production and royalties flowing to the province, the Panel approved the application. The Panel did not discuss what would happen if it was wrong and the Dina formation in this location turns out to have a limited capacity for taking injected water. Presumably it will be open to Ener T to apply to the AER to have it review its decision but it is not clear that the AER believes that Ener T has any entitlement that the AER believes that it should protect. This seems quite different from the two Kallisto decisions where the regulator was at pains to develop conditions to protect the prior natural gas storage operation from interference.

 To subscribe to ABlawg by email or RSS feed, please go to http://ablawg.ca

Follow us on Twitter @ABlawg

 

Pages