Ontario (Energy Board) v. Ontario Power Generation Inc., 2015 SCC 44

SUPREME COURT OF CANADA

Date: 20150925 Docket: 35506

Between:

Ontario Energy Board

Appellant

and

Ontario Power Generation Inc., Power Workers’ Union,

Canadian Union of Public Employees, Local 1000 and

Society of Energy Professionals

Respondents

• and -

Ontario Education Services Corporation

Intervener

Coram: McLachlin C.J. and Abella, Rothstein, Cromwell, Moldaver, Karakatsanis and Gascon JJ.

Reasons for Judgment: (paras. 1 to 121) Dissenting Reasons: (paras. 122 to 161)

Rothstein J. (McLachlin C.J. and Cromwell, Moldaver, Karakatsanis and Gascon JJ. concurring) Abella J.

Ontario (Energy Board) v. Ontario Power Generation Inc., 2015 SCC 44, [2015] 2 S.C.R. 147

Ontario Energy Board Appellant

v.

Ontario Power Generation Inc.,

Power Workers’ Union, Canadian Union

of Public Employees, Local 1000 and

Society of Energy Professionals Respondents

and

Ontario Education Services Corporation Intervener

Indexed as: Ontario (Energy Board) v. Ontario Power Generation Inc.

2015 SCC 44

File No.: 35506.

2014: December 3; 2015: September 25.

Present: McLachlin C.J. and Abella, Rothstein, Cromwell, Moldaver, Karakatsanis and Gascon JJ.

on appeal from the court of appeal for ontario

Public utilities — Electricity — Rate‑setting decision by utilities regulator — Utility seeking to recover incurred or committed compensation costs in utility rates set by Ontario Energy Board — Whether Board bound to apply particular prudence test in evaluating utility costs — Whether Board’s decision to disallow $145 million in labour compensation costs related to utility’s nuclear operations reasonable — Ontario Energy Board Act, 1998, S.O. 1998, c. 15, Sch. B, s. 78.1(5) , (6) .

Administrative law — Boards and tribunals — Appeals — Standing — Whether Ontario Energy Board acted improperly in pursuing appeal and in arguing in favour of reasonableness of its own decision — Whether Board attempted to use appeal to “bootstrap” its original decision by making additional arguments on appeal.

In Ontario, utility rates are regulated through a process by which a utility seeks approval from the Ontario Energy Board for costs the utility has incurred or expects to incur in a specified period of time. Where the Board approves of the costs, they are incorporated into utility rates such that the utility receives payment amounts to cover the approved expenditures. The Board disallowed certain payment amounts applied for by Ontario Power Generation (“OPG”) as part of its rate application covering the 2011‑2012 operating period. Specifically, the Board disallowed $145 million in labour compensation costs related to OPG’s nuclear operations on the grounds that OPG’s labour costs were out of step with those of comparable entities in the regulated power generation industry. A majority of the Ontario Divisional Court dismissed OPG’s appeal and upheld the decision of the Board. The Court of Appeal set aside the decisions of the Divisional Court and the Board and remitted the matter to the Board for redetermination in accordance with its reasons.

The crux of OPG’s argument here is that the Board is legally required to compensate OPG for all of its prudently committed or incurred costs. OPG asserts that prudence in this context has a particular methodological meaning that requires the Board to assess the reasonableness of OPG’s decision to incur or commit to costs at the time the decisions to incur or commit to the costs were made and that OPG ought to benefit from a presumption of prudence. The Board on the other hand argues that a particular prudence test methodology is not compelled by law, and that in any case the costs disallowed here were not committed nuclear compensation costs, but are better characterized as forecast costs.

OPG also raises concerns regarding the Board’s role in acting as a party on appeal from its own decision, arguing that the Board’s aggressive and adversarial defence of its decision was improper, and the Board attempted to use the appeal to bootstrap its original decision by making additional arguments on appeal. The Board argues that the structure of utilities regulation in Ontario makes it necessary and important for it to argue the merits of its decision on appeal.

Held (Abella J. dissenting): The appeal should be allowed. The decision of the Court of Appeal is set aside and the decision of the Board is reinstated.

Per McLachlin C.J. and Rothstein, Cromwell, Moldaver, Karakatsanis and Gascon JJ.: The first issue is the appropriateness of the Board’s participation in the appeal. The concerns with regard to tribunal participation on appeal from the tribunal’s own decision should not be read to establish a categorical ban. A discretionary approach provides the best means of ensuring that the principles of finality and impartiality are respected without sacrificing the ability of reviewing courts to hear useful and important information and analysis. Because of their expertise and familiarity with the relevant administrative scheme, tribunals may in many cases be well positioned to help the reviewing court reach a just outcome. Further, some cases may arise in which there is simply no other party to stand in opposition to the party challenging the tribunal decision. In a situation where no other well‑informed party stands opposed, the presence of a tribunal as an adversarial party may help the court ensure it has heard the best of both sides of a dispute. The following factors are relevant in informing the court’s exercise of its discretion: statutory provisions addressing the structure, processes and role of the particular tribunal and the mandate of the tribunal, that is, whether the function of the tribunal is to adjudicate individual conflicts between parties or whether it serves a policy‑making, regulatory or investigative role, or acts on behalf of the public interest. The importance of fairness, real and perceived, weighs more heavily against tribunal standing where the tribunal served an adjudicatory function in the proceeding. Tribunal standing is a matter to be determined by the court conducting the first‑instance review in accordance with the principled exercise of that court’s discretion. In exercising its discretion, the court is required to balance the need for fully informed adjudication against the importance of maintaining tribunal impartiality.

Consideration of these factors in the context of this case leads to the conclusion that it was not improper for the Board to participate in arguing in favour of the reasonableness of its decision on appeal. The Board was the only respondent in the initial review of its decision. It had no alternative but to step in if the decision was to be defended on the merits. Also, the Board was exercising a regulatory role by setting just and reasonable payment amounts to a utility. In this case, the Board’s participation in the instant appeal was not improper.

The issue of tribunal “bootstrapping” is closely related to the question of when it is proper for a tribunal to act as a party on appeal or judicial review of its decision. The standing issue concerns the types of argument a tribunal may make, while the bootstrapping issue concerns the content of those arguments. A tribunal engages in bootstrapping where it seeks to supplement what would otherwise be a deficient decision with new arguments on appeal. A tribunal may not defend its decision on a ground that it did not rely on in the decision under review. The principle of finality dictates that once a tribunal has decided the issues before it and provided reasons for its decision, absent a power to vary its decision or rehear the matter, it cannot use judicial review as a chance to amend, vary, qualify or supplement its reasons. While a permissive stance towards new arguments by tribunals on appeal serves the interests of justice insofar as it ensures that a reviewing court is presented with the strongest arguments in favour of both sides, to permit bootstrapping may undermine the importance of reasoned, well‑written original decisions. In this case, the Board did not impermissibly step beyond the bounds of its original decision in its arguments before the Court. The arguments raised by the Board on appeal do not amount to impermissible bootstrapping.

The merits issue concerns whether the appropriate methodology was followed by the Board in its disallowance of $145 million in labour compensation costs sought by OPG. The just‑and‑reasonable approach to recovery of the cost of services provided by a utility captures the essential balance at the heart of utilities regulation: to encourage investment in a robust utility infrastructure and to protect consumer interests, utilities must be allowed, over the long run, to earn their cost of capital, no more, no less. In order to ensure the balance between utilities’ and consumers’ interests is struck, just and reasonable rates must be those that ensure consumers are paying what the Board expects it to cost to efficiently provide the services they receive, taking account of both operating and capital costs. In that way, consumers may be assured that, overall, they are paying no more than what is necessary for the service they receive, and utilities may be assured of an opportunity to earn a fair return for providing those services.

The Ontario Energy Board Act, 1998 does not prescribe the methodology the Board must use to weigh utility and consumer interests when deciding what constitutes just and reasonable payment amounts to the utility. However, the Ontario Energy Board Act, 1998 places the burden on the applicant utility to establish that payment amounts approved by the Board are just and reasonable. It would thus seem inconsistent with the statutory scheme to presume that utility decisions to incur costs were prudent. The Board has broad discretion to determine the methods it may use to examine costs — but it cannot shift the burden of proof contrary to the statutory scheme.

The issue is whether the Board was bound to use a no-hindsight, presumption of prudence test to determine whether labour compensation costs were just and reasonable. The prudent investment test, or prudence review, is a valid and widely accepted tool that regulators may use when assessing whether payments to a utility would be just and reasonable. However, there is no support in the statutory scheme for the notion that the Board should be required as a matter of law, under the Ontario Energy Board Act, 1998 to apply the prudence test such that the mere decision not to apply it when considering committed costs would render its decision on payment amounts unreasonable. Where a statute requires only that the regulator set “just and reasonable” payments, as the Ontario Energy Board Act, 1998 does in Ontario, the regulator may make use of a variety of analytical tools in assessing the justness and reasonableness of a utility’s proposed payment amounts. This is particularly so where, as here, the regulator has been given express discretion over the methodology to be used in setting payment amounts.

Where the regulator has discretion over its methodological approach, understanding whether the costs at issue are “forecast” or “committed” may be helpful in reviewing the reasonableness of a regulator’s choice of methodology. Here, the labour compensation costs which led to the $145 million disallowance are best understood as partly committed costs and partly costs subject to management discretion. They are partly committed because they resulted from collective agreements entered into between OPG and two of its unions, and partly subject to management discretion because OPG retained some flexibility to manage total staffing levels in light of, among other things, projected attrition of the workforce. It is not reasonable to treat these costs as entirely forecast. However, the Board was not bound to apply a particular prudence test in evaluating these costs. It is not necessarily unreasonable, in light of the particular regulatory structure established by the Ontario Energy Board Act, 1998 , for the Board to evaluate committed costs using a method other than a no‑hindsight prudence review. Applying a presumption of prudence would have conflicted with the burden of proof in the Ontario Energy Board Act, 1998 and would therefore not have been reasonable. The question of whether it was reasonable to assess a particular cost using hindsight should turn instead on the circumstances of that cost.

In this case, the nature of the disputed costs and the environment in which they arose provide a sufficient basis to find that the Board did not act unreasonably in not applying the prudent investment test in determining whether it would be just and reasonable to compensate OPG for these costs and disallowing them. Since the costs at issue are operating costs, there is little danger that a disallowance of these costs will have a chilling effect on OPG’s willingness to incur operating costs in the future, because costs of the type disallowed here are an inescapable element of operating a utility. Further, the costs at issue arise in the context of an ongoing repeat‑player relationship between OPG and its employees. Such a context supports the reasonableness of a regulator’s decision to weigh all evidence it finds relevant in striking a just and reasonable balance between the utility and consumers, rather than confining itself to a no‑hindsight approach. There is no dispute that collective agreements are “immutable” between employees and the utility. However, if the legislature had intended for costs under collective agreements to also be inevitably imposed on consumers, it would not have seen fit to grant the Board oversight of utility compensation costs. The Board’s decision in no way purports to force OPG to break its contractual commitments to unionized employees. It was not unreasonable for the Board to adopt a mixed approach that did not rely on quantifying the exact share of compensation costs that fell into the forecast and committed categories. Such an approach represents an exercise of the Board’s methodological discretion in addressing a challenging issue where these costs did not fit easily into one category or the other.

The Board’s disallowance may have adversely impacted OPG’s ability to earn its cost of capital in the short run. Nevertheless, the disallowance was intended to send a clear signal that OPG must take responsibility for improving its performance. Such a signal may, in the short run, provide the necessary impetus for OPG to bring its compensation costs in line with what, in the Board’s opinion, consumers should justly expect to pay for an efficiently provided service. Sending such a signal is consistent with the Board’s market proxy role and its objectives under s. 1 of the Ontario Energy Board Act, 1998 .

Per Abella J. (dissenting): The Board’s decision was unreasonable because the Board failed to apply the methodology set out for itself for evaluating just and reasonable payment amounts. It both ignored the legally binding nature of the collective agreements between Ontario Power Generation and the unions and failed to distinguish between committed compensation costs and those that were reducible.

The Board stated in its reasons that it would use two kinds of review in order to determine just and reasonable payment amounts. As to “forecast costs”, that is, those over which a utility retains discretion and can still be reduced or avoided, the Board explained that it would review such costs using a wide range of evidence, and that the onus would be on the utility to demonstrate that its forecast costs were reasonable. A different approach, however, would be applied to those costs the company could not “take action to reduce”. These costs, sometimes called “committed costs”, represent binding commitments that leave a utility with no discretion about whether to make the payment. The Board explained that it would evaluate these costs using a “prudence review”. The application of a prudence review does not shield these costs from scrutiny, but it does include a presumption that the costs were prudently incurred.

Rather than apply the methodology it set out for itself, however, the Board assessed all compensation costs in Ontario Power Generation’s collective agreements as adjustable forecast costs, without determining whether any of them were costs for which there is no opportunity for the company to take action to reduce. The Board’s failure to separately assess the compensation costs committed as a result of the collective agreements from other compensation costs, ignored not only its own methodological template, but labour law as well.

The compensation costs for approximately 90 per cent of Ontario Power Generation’s regulated workforce were established through legally binding collective agreements which obligated the utility to pay fixed levels of compensation, regulated staffing levels, and provided unionized employees with employment security. The obligations contained in these collective agreements were immutable and legally binding commitments. The agreements therefore did not just leave the utility with limited flexibility regarding overall compensation or staffing levels, they made it illegal for the utility to alter the compensation and staffing levels of 90 per cent of its regulated workforce in a manner that was inconsistent with its commitments under the agreements.

The Board, however, applying the methodology it said it would use for the utility’s forecast costs, put the onus on Ontario Power Generation to prove the reasonableness of all its compensation costs and concluded that it had failed to provide compelling evidence or documentation or analysis to justify compensation levels. Had the Board used the approach it said it would use for costs the company had no opportunity to reduce, it would have used an after‑the‑fact prudence review, with a rebuttable presumption that the utility’s expenditures were reasonable.

It may well be that Ontario Power Generation has the ability to manage some staffing levels through attrition or other mechanisms that did not breach the utility’s commitments under its collective agreements, and that these costs may therefore properly be characterized as forecast costs. But no factual findings were made by the Board about the extent of any such flexibility. There is in fact no evidence in the record, nor any evidence cited in the Board’s decision, setting out what proportion of Ontario Power Generation’s compensation costs were fixed and what proportion remained subject to the utility’s discretion. Given that collective agreements are legally binding, it was unreasonable for the Board to assume that Ontario Power Generation could reduce the costs fixed by these contracts in the absence of any evidence to that effect.

Selecting a test which is more likely to confirm the Board’s assumption that collectively‑bargained costs are excessive, misconceives the point of the exercise, namely, to determine whether those costs were in fact excessive. Blaming collective bargaining for what are assumed to be excessive costs, imposes the appearance of an ideologically‑driven conclusion on what is intended to be a principled methodology based on a distinction between committed and forecast costs, not between costs which are collectively bargained and those which are not. While the Board has wide discretion to fix payment amounts that are just and reasonable and, subject to certain limitations, to establish the methodology used to determine such amounts, once the Board establishes a methodology, it is, at the very least, required to faithfully apply it.

Absent methodological clarity and predictability, Ontario Power Generation would be unable to know how to determine what expenditures and investments to make and how to present them to the Board for review. Wandering sporadically from approach to approach, or failing to apply the methodology it declares itself to be following, creates uncertainty and leads, inevitably, to needlessly wasting public time and resources in constantly having to anticipate and respond to moving regulatory targets. Whether or not one can fault the Board for failing to use a particular methodology, what the Board can unquestionably be analytically faulted for, is evaluating all compensation costs fixed by collective agreements as being amenable to adjustment. Treating these compensation costs as reducible was unreasonable.

The appeal should accordingly be dismissed, the Board’s decision set aside, and the matter remitted to the Board for reconsideration.

Cases Cited

By Rothstein J.

Considered: Enbridge Gas Distribution Inc. v. Ontario Energy Board (2006), 2006 10734 (ON CA) , 210 O.A.C. 4; Northwestern Utilities Ltd. v. City of Edmonton , 1978 17 (SCC) , [1979] 1 S.C.R. 684; referred to: Toronto Hydro‑Electric System Ltd. v. Ontario (Energy Board) , 2010 ONCA 284 , 99 O.R. (3d) 481; Northwestern Utilities Ltd. v. City of Edmonton , 1929 39 (SCC) , [1929] S.C.R. 186; TransCanada Pipelines Ltd. v. National Energy Board , 2004 FCA 149 , 319 N.R. 171; Ontario Power Generation Inc. (Re) , EB‑2007‑0905, November 3, 2008 (online: http://www.ontarioenergyboard.ca/); CAIMAW v. Paccar of Canada Ltd. , 1989 49 (SCC) , [1989] 2 S.C.R. 983; B.C.G.E.U. v. Indust. Rel. Council (1988), 1988 2812 (BC CA) , 26 B.C.L.R. (2d) 145; McLean v. British Columbia (Securities Commission) , 2013 SCC 67 , [2013] 3 S.C.R. 895; Ellis‑Don Ltd. v. Ontario (Labour Relations Board) , 2001 SCC 4 , [2001] 1 S.C.R. 221; Tremblay v. Quebec (Commission des affaires sociales) , 1992 1135 (SCC) , [1992] 1 S.C.R. 952; Ontario (Children’s Lawyer) v. Ontario (Information and Privacy Commissioner) (2005), 2005 11786 (ON CA) , 75 O.R. (3d) 309; Canada (Attorney General) v. Quadrini , 2010 FCA 246 , [2012] 2 F.C.R. 3; Leon’s Furniture Ltd. v. Information and Privacy Commissioner (Alta.) , 2011 ABCA 94 , 502 A.R. 110; Henthorne v. British Columbia Ferry Services Inc. , 2011 BCCA 476 , 344 D.L.R. (4th) 292; United Brotherhood of Carpenters and Joiners of America, Local 1386 v. Bransen Construction Ltd. , 2002 NBCA 27 , 249 N.B.R. (2d) 93; Chandler v. Alberta Association of Architects , 1989 41 (SCC) , [1989] 2 S.C.R. 848; Dunsmuir v. New Brunswick , 2008 SCC 9 , [2008] 1 S.C.R. 190; Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association , 2011 SCC 61 , [2011] 3 S.C.R. 654; Tervita Corp. v. Canada (Commissioner of Competition) , 2015 SCC 3 , [2015] 1 S.C.R. 161; Bell Canada v. Bell Aliant Regional Communications , 2009 SCC 40 , [2009] 2 S.C.R. 764; Re General Increase in Freight Rates (1954), 76 C.R.T.C. 12 ; ATCO Gas and Pipelines Ltd. v. Alberta (Energy and Utilities Board) , 2006 SCC 4 , [2006] 1 S.C.R. 140; State of Missouri ex rel. Southwestern Bell Telephone Co. v. Public Service Commission of Missouri , 262 U.S. 276 (1923); Duquesne Light Co. v. Barasch , 488 U.S. 299 (1989); U.S. West Communications, Inc. v. Public Service Commission of Utah , 901 P.2d 270 (1995); British Columbia Electric Railway Co. v. Public Utilities Commission of British Columbia , 1960 44 (SCC) , [1960] S.C.R. 837; Nova Scotia Power Inc., Re , 2005 NSUARB 27 ; Nova Scotia Power Inc. (Re) , 2012 NSUARB 227 .

By Abella J. (dissenting)

Verizon Communications Inc. v. Federal Communications Commission , 535 U.S. 467 (2002); Northwestern Utilities Ltd. v. City of Edmonton , 1929 39 (SCC) , [1929] S.C.R. 186; State of Missouri ex rel. Southwestern Bell Telephone Co. v. Public Service Commission of Missouri , 262 U.S. 276 (1923); Enersource Hydro Mississauga Inc. (Re) , 2012 LNONOEB 373 (QL); Enbridge Gas Distribution Inc. (Re) , 2002 LNONOEB 4 (QL); Enbridge Gas Distribution Inc. v. Ontario Energy Board (2006), 2006 10734 (ON CA) , 210 O.A.C. 4; Ontario Power Generation v. Society of Energy Professionals , [2011] O.L.A.A. No. 117 (QL); TransCanada Pipelines Ltd. v. National Energy Board , 2004 FCA 149 , 319 N.R. 171.

Statutes and Regulations Cited

Labour Relations Act, 1995 , S.O. 1995, c. 1, Sch. A, ss. 56, 69 .

Nuclear Safety and Control Act , S.C. 1997, c. 9.

Ontario Energy Board Act, 1998 , S.O. 1998, c. 15, Sch. B, ss. 1, 33(3) , 78.1 .

Payments Under Section 78.1 of the Act , O. Reg. 53/05, ss. 3, 6 .

Public Utilities Act , R.S.B.C. 1948, c. 277 [rep. 1973, c. 29, s. 187], s. 16(1)( b ).

Authors Cited

Burns, Robert E., et al. The Prudent Investment Test in the 1980s , report NRRI‑84‑16. Columbus, Ohio: National Regulatory Research Institute, April 1985.

Chaykowski, Richard P. An Assessment of the Industrial Relations Context and Outcomes at OPG , file No. EB‑2013‑0321, exhibit F4‑03‑01, attachment 1, September 2013 (online: http://www.opg.com/about/regulatory-affairs/Documents/2014-2015/F4‑03‑01_Attachment%201.pdf).

Clark, Ron W., Scott A. Stoll and Fred D. Cass. Ontario Energy Law: Electricity . Markham, Ont.: LexisNexis, 2012.

Falzon, Frank A. V. “Tribunal Standing on Judicial Review” (2008), 21 C.J.A.L.P. 21.

Jacobs, Laverne A., and Thomas S. Kuttner. “Discovering What Tribunals Do: Tribunal Standing Before the Courts” (2002), 81 Can. Bar Rev. 616.

Kahn, Jonathan. “Keep Hope Alive: Updating the Prudent Investment Standard for Allocating Nuclear Plant Cancellation Costs” (2010), 22 Fordham Envtl. L. Rev. 43.

Mullan, David. “Administrative Law and Energy Regulation”, in Gordon Kaiser and Bob Heggie, eds., Energy Law and Policy . Toronto: Carswell, 2011, 35.

Ontario. Office of the Auditor General of Ontario. 2011 Annual Report . Toronto: Queen’s Printer, 2011.

Reid, Laurie, and John Todd. “New Developments in Rate Design for Electricity Distributors”, in Gordon Kaiser and Bob Heggie, eds., Energy Law and Policy . Toronto: Carswell, 2011, 519.

Semple, Noel. “The Case for Tribunal Standing in Canada” (2007), 20 C.J.A.L.P. 305.

APPEAL from a judgment of the Ontario Court of Appeal (Rosenberg, Goudge and Blair JJ.A.), 2013 ONCA 359 , 116 O.R. (3d) 793, 365 D.L.R. (4th) 247, 307 O.A.C. 109, [2013] O.J. No. 3917 (QL), 2013 CarswellOnt 9792 (WL Can.), setting aside a decision of the Divisional Court (Aitken, Swinton and Hoy JJ.), 2012 ONSC 729 , 109 O.R. (3d) 576, 347 D.L.R. (4th) 355, [2012] O.J. No. 862 (QL), 2012 CarswellOnt 2710 (WL Can.), and setting aside a decision of the Ontario Energy Board, EB‑2010‑0008, March 10, 2011 (online: http://www.ontarioenergyboard.ca/), 2011 LNONOEB 57 (QL), 2011 CarswellOnt 3723 (WL Can.) . Appeal allowed, Abella J. dissenting.

Glenn Zacher , Patrick Duffy and James Wilson , for the appellant.

John B. Laskin , Crawford Smith , Myriam Seers and Carlton Mathias , for the respondent Ontario Power Generation Inc .

Richard P. Stephenson and Emily Lawrence , for the respondent the Power Workers’ Union, Canadian Union of Public Employees, Local 1000 .

Paul J. J. Cavalluzzo and Amanda Darrach , for the respondent the Society of Energy Professionals .

Mark Rubenstein , for the intervener.

The judgment of McLachlin C.J. and Rothstein, Cromwell, Moldaver, Karakatsanis and Gascon JJ. was delivered by

[ 1 ] Rothstein J. — In Ontario, utility rates are regulated through a process by which a utility seeks approval from the Ontario Energy Board (“Board”) for costs the utility has incurred or expects to incur in a specified period of time. Where the Board approves of costs, they are incorporated into utility rates such that the utility receives payment amounts to cover the approved expenditures. This case concerns the decision of the Board to disallow certain payment amounts applied for by Ontario Power Generation Inc. (“OPG”) as part of its rate application covering the 2011-2012 operating period. Specifically, the Board disallowed $145 million in labour compensation costs related to OPG’s nuclear operations on the grounds that OPG’s labour costs were out of step with those of comparable entities in the regulated power generation industry.

[ 2 ] OPG appealed the Board’s decision to the Ontario Divisional Court. A majority of the court dismissed the appeal and upheld the decision of the Board. OPG then appealed that decision to the Ontario Court of Appeal, which set aside the decisions of the Divisional Court and the Board and remitted the matter to the Board for redetermination in accordance with its reasons. The Board now appeals to this Court.

[ 3 ] OPG asserts that the Board’s decision to disallow these labour compensation costs was unreasonable. The crux of OPG’s argument is that the Board is legally required to compensate OPG for all of its prudently committed or incurred costs. OPG asserts that prudence in this context has a particular methodological meaning that requires the Board to assess the reasonableness of OPG’s decisions to incur or commit to costs at the time the decisions to incur or commit to the costs were made and that OPG ought to benefit from a presumption of prudence. Because the Board did not employ this prudence methodology, OPG argues that its decision was unreasonable.

[ 4 ] The Board argues that a particular “prudence test” methodology is not compelled by law, and that in any case the costs disallowed here were not “committed” nuclear compensation costs, but are better characterized as forecast costs.

[ 5 ] OPG also raises concerns regarding the Board’s role in acting as a party on appeal from its own decision. OPG argues that in this case, the Board’s aggressive and adversarial defence of its original decision was improper, and that the Board attempted to use the appeal to “bootstrap” its original decision by making additional arguments on appeal.

[ 6 ] The Board asserts that the scope of its authority to argue on appeal was settled when it was granted full party rights in connection with the granting of leave by this Court. Alternatively, the Board argues that the structure of utilities regulation in Ontario makes it necessary and important for it to argue the merits of its decisions on appeal.

[ 7 ] In my opinion, the labour compensation costs which led to the $145 million disallowance are best understood as partly committed costs and partly costs subject to management discretion. They are partly committed because they resulted from collective agreements entered into between OPG and two of its unions, and partly subject to management discretion because OPG retained some flexibility to manage total staffing levels in light of, among other things, projected attrition of the workforce. It is not reasonable to treat these costs as entirely forecast. However, I do not agree with OPG that the Board was bound to apply a particular prudence test in evaluating these costs. The Ontario Energy Board Act, 1998 , S.O. 1998, c. 15, Sch. B, and associated regulations give the Board broad latitude to determine the methodology it uses in assessing utility costs, subject to the Board’s ultimate duty to ensure that payment amounts it orders be just and reasonable to both the utility and consumers.

[ 8 ] In this case, the nature of the disputed costs and the environment in which they arose provide a sufficient basis to find that the Board did not act unreasonably in disallowing the costs.

[ 9 ] Regarding the Board’s role on appeal, I do not find that the Board acted improperly in arguing the merits of this case, nor do I find that the arguments raised on appeal amount to impermissible “bootstrapping”.

[ 10 ] Accordingly, I would allow the appeal, set aside the decision of the Court of Appeal, and reinstate the decision of the Board.

I. Regulatory Framework

[ 11 ] The Ontario Energy Board Act, 1998 establishes the Board as a regulatory body with authority to oversee, among other things, electricity generation in the province of Ontario. Section 1 sets out the objectives of the Board in regulating electricity, which include:

1. (1) . . .

2. To protect the interests of consumers with respect to prices and the adequacy, reliability and quality of electricity service.

3. To promote economic efficiency and cost effectiveness in the generation, transmission, distribution, sale and demand management of electricity and to facilitate the maintenance of a financially viable electricity industry.

Accordingly, the Board must ensure that it regulates with an eye to balancing both consumer interests and the efficiency and financial viability of the electricity industry. The Board’s role has also been described as that of a “market proxy”: 2012 ONSC 729 , 109 O.R. (3d) 576, at para. 54 ; 2013 ONCA 359 , 116 O.R. (3d) 793, at para. 38 . In this sense, the Board’s role is to emulate as best as possible the forces to which a utility would be subject in a competitive landscape: Toronto Hydro-Electric System Ltd. v. Ontario (Energy Board) , 2010 ONCA 284 , 99 O.R. (3d) 481, at para. 48 .

[ 12 ] One of the Board’s most powerful tools to achieve its objectives is its authority to fix the amount of payments utilities receive in exchange for the provision of service. Section 78.1(5) of the Ontario Energy Board Act, 1998 provides in relevant part:

(5) The Board may fix such other payment amounts as it finds to be just and reasonable,

(a) on an application for an order under this section, if the Board is not satisfied that the amount applied for is just and reasonable; . . .

[ 13 ] Section 78.1(6) provides: “. . . the burden of proof is on the applicant in an application made under this section”.

[ 14 ] As I read these provisions, the utility applies for payment amounts for a future period (called the “test period”). The Board will accept the payment amounts applied for unless the Board is not satisfied that the amounts are just and reasonable. Where the Board is not satisfied, s. 78.1(5) empowers it to fix other payment amounts which it finds to be just and reasonable.

[ 15 ] This Court has had the occasion to consider the meaning of similar statutory language in Northwestern Utilities Ltd. v. City of Edmonton , 1929 39 (SCC) , [1929] S.C.R. 186. In that case, the Court held that “fair and reasonable” rates were those “which, under the circumstances, would be fair to the consumer on the one hand, and which, on the other hand, would secure to the company a fair return for the capital invested” (pp. 192-93).

[ 16 ] This means that the utility must, over the long run, be given the opportunity to recover, through the rates it is permitted to charge, its operating and capital costs (“capital costs” in this sense refers to all costs associated with the utility’s invested capital). This case is concerned primarily with operating costs. If recovery of operating costs is not permitted, the utility will not earn its cost of capital, which represents the amount investors require by way of a return on their investment in order to justify an investment in the utility. The required return is one that is equivalent to what they could earn from an investment of comparable risk. Over the long run, unless a regulated utility is allowed to earn its cost of capital, further investment will be discouraged and it will be unable to expand its operations or even maintain existing ones. This will harm not only its shareholders, but also its customers: TransCanada Pipelines Ltd. v. National Energy Board , 2004 FCA 149 , 319 N.R. 171.

[ 17 ] This of course does not mean that the Board must accept every cost that is submitted by the utility, nor does it mean that the rate of return to equity investors is guaranteed. In the short run, return on equity may vary, for example if electricity consumption by the utility’s customers is higher or lower than predicted. Similarly, a disallowance of any operating costs to which the utility has committed itself will negatively impact the return to equity investors. I do not intend to enter into a detailed analysis of how the cost of equity capital should be treated by utility regulators, but merely to observe that any disallowance of costs to which a utility has committed itself has an effect on equity investor returns. This effect must be carefully considered in light of the long-run necessity that utilities be able to attract investors and retain earnings in order to survive and operate efficiently and effectively, in accordance with the statutory objectives of the Board in regulating electricity in Ontario.

[ 18 ] As noted above, the burden is on the utility to satisfy the Board that the payment amounts it applies for are just and reasonable. If it fails to do so, the Board may disallow the portion of the application that it finds is not for amounts that are just and reasonable.

[ 19 ] Where applied-for operating costs are disallowed, the utility, if it is able to do so, may forego the expenditure of such costs. Where the expenditure cannot be foregone, the shareholders of the utility will have to absorb the reduction in the form of receiving less than their anticipated rate of return on their investment, i.e. the utility’s cost of equity capital. In such circumstances it will be the management of the utility that will be responsible in the future for bringing its costs into line with what the Board considers just and reasonable.

[ 20 ] In order to ensure that the balance between utilities’ and consumers’ interests is struck, just and reasonable rates must be those that ensure consumers are paying what the Board expects it to cost to efficiently provide the services they receive, taking account of both operating and capital costs. In that way, consumers may be assured that, overall, they are paying no more than what is necessary for the service they receive, and utilities may be assured of an opportunity to earn a fair return for providing those services.

II. Facts

[ 21 ] OPG is Ontario’s largest energy generator, and is subject to rate regulation by the Board. OPG came into being in 1999 as one of the successor corporations to Ontario Hydro. It operates Board-regulated nuclear and hydroelectric facilities that generate approximately half of Ontario’s electricity. Its sole shareholder is the Province of Ontario.

[ 22 ] It employs approximately 10,000 people in connection with its regulated facilities, 95 percent of whom work in its nuclear business. Approximately 90 percent of its employees in its regulated businesses are unionized, with approximately two thirds of unionized employees represented by the Power Workers’ Union, Canadian Union of Public Employees, Local 1000 (“PWU”), and one third represented by the Society of Energy Professionals (“Society”).

[ 23 ] Since early in its existence as an independent utility, OPG has been aware of the importance of improving its corporate performance. As part of a general effort to improve its business, OPG undertook efforts to benchmark its nuclear performance against comparable power plants around the world. In a memorandum of agreement (“MOA”) with the Province of Ontario dated August 17, 2005, OPG committed to the following:

OPG will seek continuous improvement in its nuclear generation business and internal services. OPG will benchmark its performance in these areas against CANDU nuclear plants worldwide as well as against the top quartile of private and publicly-owned nuclear electricity generators in North America. OPG’s top operational priority will be to improve the operation of its existing nuclear fleet.

(A.R., vol. III, at p. 215)

[ 24 ] As part of OPG’s first-ever rate application with the Board in 2007, for a test period covering the years 2008 and 2009, OPG sought approval for a $6.4 billion “revenue requirement”; this term refers to “the total revenue that is required by the company to pay all of its allowable expenses and also to recover all costs associated with its invested capital”: L. Reid and J. Todd, “New Developments in Rate Design for Electricity Distributors”, in G. Kaiser and B. Heggie, eds., Energy Law and Policy (2011), 519, at p. 521. This constituted an increase of $1 billion over the revenue requirement that it had sought and was granted under the regulatory scheme in place prior to the Board’s assumption of regulatory authority over OPG: EB-2007-0905, Decision with Reasons, November 3, 2008 (“Board 2008-2009 Decision”) (online), at pp. 5-6.

[ 25 ] The Board found that OPG was not meeting the nuclear performance expectations of its sole shareholder and that it had done little to conduct benchmarking of its performance against that of its peers, despite its commitment to do so dating back to 2005. Indeed, the only evidence of benchmarking that OPG submitted as part of its rate application was a 2006 report from Navigant Consulting, Inc. (“Navigant Report”), which found that OPG was overstaffed by 12 percent in comparison to its peers. The Board found that OPG had not acted on the recommendations of the Navigant Report and had not commissioned subsequent benchmarking studies to assess its performance (Board 2008-2009 Decision, at pp. 27 and 30). The Board also found that operating costs at OPG’s Pickering nuclear facilities were “far above industry averages” (p. 29). The Board thus disallowed $35 million of OPG’s proposed revenue requirement and directed OPG to prepare benchmarking studies for use in future applications (p. 31).

[ 26 ] In explaining the importance of benchmarking, the Board stated: “The reason why the MOA emphasized benchmarking was because such studies can and do shine a light on inefficiencies and lack of productivity improvement” (Board 2008-2009 Decision, at p. 30).

[ 27 ] On May 5, 2010, shortly before OPG was set to file its second rate application, which is the subject of this appeal, the Ontario Minister of Energy and Infrastructure wrote to the President and CEO of OPG to ensure that OPG would demonstrate in its upcoming rate application “concerted efforts to identify cost saving opportunities and focus [its] forthcoming rate application on those items that are essential to the safe and reliable operation of [its] existing assets and projects already under development” ( A.R., vol. IV, at p. 38 ).

[ 28 ] On May 26, 2010, OPG filed its payment amounts application for the 2011-2012 test period. As part of its evidence before the Board, OPG submitted two reports by ScottMadden Inc., a general management consulting firm specializing in benchmarking and business planning for nuclear facilities. The Phase 1 report compared OPG’s nuclear operational and financial performance against that of external peers using industry performance metrics. The Phase 2 final report discussed performance improvement targets with the intent of improving OPG’s nuclear business. OPG collaborated with ScottMadden on the Phase 1 and 2 reports, which were released on July 2, 2009 and September 11, 2009, respectively.

[ 29 ] OPG’s rate application pertained to a test period beginning on January 1, 2011 and ending on December 31, 2012. OPG sought approval of a $6.9 billion revenue requirement, which represented an increase of 6.2 percent over OPG’s then-current revenue based on the preceding year’s approved utility rates. Of the $6.9 billion revenue requirement sought by OPG, $2.8 billion pertained to compensation costs, of which approximately $2.4 billion concerned OPG’s nuclear business.

[ 30 ] A substantial portion of OPG’s wage and compensation expenses was fixed by OPG’s collective agreements with the unions, PWU and the Society. At the time of its application, OPG was party to a collective agreement with PWU, effective from April 2009 through March 2012, while its collective agreement with the Society expired on December 31, 2010. These collective agreements provided annual wage increases between 2 percent and 3 percent. OPG forecast an additional 1 percent increase for step progressions and promotions of unionized staff. Following the Board’s hearing in this case, an interest arbitrator ordered a new collective agreement between OPG and the Society, effective February 3, 2011. This collective agreement provided wage increases that varied between 1 percent and 3 percent.

III. Judicial History

A. Ontario Energy Board: 2011 LNONOEB 57 (QL) (“Board Decision”)

[ 31 ] In its decision concerning OPG’s rate application for the 2011-2012 test period, the Board stated that it enjoyed broad discretion pursuant to Ontario Regulation 53/05 ( Payments Under Section 78.1 of the Act ) and s. 78.1 of the Ontario Energy Board Act, 1998 to “adopt the mechanisms it judges appropriate in setting just and reasonable rates” (para. 73). The Board recognized that different tests could apply depending on whether its analysis concerned the recovery of forecast costs or an after-the-fact review of costs already incurred. In this rate application, it was appropriate to take into consideration all evidence that the Board deemed relevant to assess the reasonableness of OPG’s revenue requirement.

[ 32 ] The Board rejected OPG’s proposed revenue requirement of $6.9 billion, reducing it by $145 million over the test period “to send a clear signal that OPG must take responsibility for improving its performance” (para. 350). Key to its disallowance was the Board’s finding that OPG was overstaffed and that its compensation levels were excessive.

[ 33 ] Regarding the number of staff, the Board pointed out that a benchmarking study commissioned by OPG itself, the ScottMadden Phase 2 final report, suggested that certain staff positions could be reduced or eliminated altogether. The Board suggested that OPG could review its organizational structure and reassign or eliminate positions in the coming years, as 20 percent to 25 percent of its staff were set to retire between 2010 and 2014 and it was possible to make greater use of external contractors. Regarding compensation, the Board found that OPG had not submitted compelling evidence justifying the benchmarking of its salaries of non-management employees to the 75th percentile of a survey of industry salaries conducted by Towers Perrin. Instead, the Board considered the proper benchmark to be the 50th percentile, the same percentile against which OPG benchmarks management compensation. In determining the appropriate disallowance, the Board acknowledged that OPG may not have been able to achieve the full $145 million in savings for the test period through the reduction of compensation levels alone because of its collective agreements with the unions.

B. Ontario Superior Court of Justice, Divisional Court: 2012 ONSC 729 , 109 O.R. (3d) 576

[ 34 ] OPG appealed the Board Decision on the basis that it was unreasonable and that the reasons provided were inadequate. OPG argued that the Board should have conducted a prudent investment test — that is, it should have restricted its review of compensation costs to a consideration of whether the collective agreements that prescribed the compensation costs were prudent at the time they were entered into. OPG also argued that the Board should have presumed that the costs were prudent.

[ 35 ] The panel of three Divisional Court judges was split. Justice Hoy (as she then was), for the majority, found the Board Decision reasonable because management had the ability to reduce total compensation costs in the future within the framework of the collective agreement. Applying a strict prudent investment test would not permit the Board to fulfill its statutory objective of promoting cost effectiveness in the generation of electricity. It was particularly important for the Board to exercise its authority to set just and reasonable rates given the “double monopoly” dynamic at play:

The collective agreements were concluded between a regulated monopoly, which passes costs on to consumers, not a competitive enterprise, and two unions which account for approximately 90 per cent of the employees and amount to a near, second monopoly, based on terms inherited from Ontario Hydro and in face of the reality that running a nuclear operation without the employees would be extremely difficult. [para. 54]

[ 36 ] Justice Aitken dissented, finding that,

to the extent that [nuclear compensation] costs were predetermined, in the sense that they were locked in as a result of collective agreements entered prior to the date of the application and the test period, OPG only had to prove their prudence or reasonableness based on the circumstances that were known or that reasonably could have been anticipated at the time the decision to enter those collective agreements was made. [para. 83]

She would have held that the Board’s failure to undertake a separate and explicit prudence review for the committed portion of nuclear compensation costs, coupled with its consideration of hindsight factors in assessing the reasonableness of these costs, rendered the Board Decision unreasonable.

C. Ontario Court of Appeal: 2013 ONCA 359 , 116 O.R. (3d) 793

[ 37 ] The Ontario Court of Appeal reversed the Divisional Court’s decision and remitted the case to the Board. The court drew a distinction between forecast costs and committed costs, with committed costs being those that the utility “is committed to pay in [the test period]” and that “cannot be managed or reduced by the utility in that time frame, usually because of contractual obligations” (para. 29). Although costs may not require actual payment until the future, as in this case, costs that have been “contractually incurred to be paid over the time frame are nonetheless committed even though they have not yet been paid” (para. 29). When reviewing such costs, the court held that the Board must undertake a prudence review as described in Enbridge Gas Distribution Inc. v. Ontario Energy Board (2006), 2006 10734 (ON CA) , 210 O.A.C. 4 (paras. 15-16 ). By failing to follow this jurisprudence and by requiring that OPG “manage costs that, by law, it cannot manage”, the Board acted unreasonably (para. 37).

IV. Issues

[ 38 ] The Board raises two issues on appeal:

1. What is the appropriate standard of review?

2. Was the Board’s decision to disallow $145 million of OPG’s revenue requirement reasonable?

[ 39 ] Before this Court, OPG has argued that the Board stepped beyond the appropriate role of a tribunal in an appeal from its own decision, which raises the following additional issue:

3. Did the Board act impermissibly in pursuing its appeal in this case?

V. Analysis

[ 40 ] It is logical to begin by considering the appropriateness of the Board’s participation in the appeal. I will next consider the appropriate standard of review, and then the merits issue of whether the Board’s decision in this case was reasonable.

A. The Appropriate Role of the Board in This Appeal

(1) Tribunal Standing

[ 41 ] In Northwestern Utilities Ltd. v. City of Edmonton , 1978 17 (SCC) , [1979] 1 S.C.R. 684 (“ Northwestern Utilities ”), per Estey J., this Court first discussed how an administrative decision-maker’s participation in the appeal or review of its own decisions may give rise to concerns over tribunal impartiality. Estey J. noted that “active and even aggressive participation can have no other effect than to discredit the impartiality of an administrative tribunal either in the case where the matter is referred back to it, or in future proceedings involving similar interests and issues or the same parties” (p. 709). He further observed that tribunals already receive an opportunity to make their views clear in their original decisions: “. . . it abuses one’s notion of propriety to countenance its participation as a full-fledged litigant in this Court” (p. 709) .

[ 42 ] The Court in Northwestern Utilities ultimately held that the Alberta Public Utilities Board — which, like the Ontario Energy Board, had a statutory right to be heard on judicial appeal (see Ontario Energy Board Act, 1998 , s. 33(3)) — was limited in the scope of the submissions it could make. Specifically, Estey J. observed that

[i]t has been the policy in this Court to limit the role of an administrative tribunal whose decision is at issue before the Court, even where the right to appear is given by statute, to an explanatory role with reference to the record before the Board and to the making of representations relating to jurisdiction. [p. 709]

[ 43 ] This Court further considered the issue of agency standing in CAIMAW v. Paccar of Canada Ltd. , 1989 49 (SCC) , [1989] 2 S.C.R. 983, which involved judicial review of a British Columbia Labour Relations Board decision. Though a majority of the judges hearing the case did not endorse a particular approach to the issue, La Forest J., Dickson C.J. concurring, accepted that a tribunal had standing to explain the record and advance its view of the appropriate standard of review and, additionally, to argue that its decision was reasonable.

[ 44 ] This finding was supported by the need to make sure the Court’s decision on review of the tribunal’s decision was fully informed. La Forest J. cited B.C.G.E.U. v. Indust. Rel. Council (1988), 1988 2812 (BC CA) , 26 B.C.L.R. (2d) 145 (C.A.), at p. 153, for the proposition that the tribunal is the party best equipped to draw the Court’s attention to

those considerations, rooted in the specialized jurisdiction or expertise of the tribunal, which may render reasonable what would otherwise appear unreasonable to someone not versed in the intricacies of the specialized area.

( Paccar , at p. 1016)

La Forest J. found, however, that the tribunal could not go so far as to argue that its decision was correct (p. 1017). Though La Forest J. did not command a majority, L’Heureux-Dubé J. also commented on tribunal standing in her dissent, and agreed with the substance of La Forest J.’s analysis (p. 1026).

[ 45 ] Trial and appellate courts have struggled to reconcile this Court’s statements in Northwestern Utilities and Paccar . Indeed, while this Court has never expressly overturned Northwestern Utilities , on some occasions, it has permitted tribunals to participate as full parties without comment: see, e.g., McLean v. British Columbia (Securities Commission) , 2013 SCC 67 , [2013] 3 S.C.R. 895; Ellis-Don Ltd. v. Ontario (Labour Relations Board) , 2001 SCC 4 , [2001] 1 S.C.R. 221; Tremblay v. Quebec (Commission des affaires sociales) , 1992 1135 (SCC) , [1992] 1 S.C.R. 952; see also Ontario (Children’s Lawyer) v. Ontario (Information and Privacy Commissioner) (2005), 2005 11786 (ON CA) , 75 O.R. (3d) 309 (C.A.) (“ Goodis ”), at para. 24 .

[ 46 ] A number of appellate decisions have grappled with this issue and “for the most part now display a more relaxed attitude in allowing tribunals to participate in judicial review proceedings or statutory appeals in which their decisions were subject to attack”: D. Mullan, “Administrative Law and Energy Regulation”, in G. Kaiser and B. Heggie, 35, at p. 51. A review of three appellate decisions suffices to establish the rationale behind this shift.

[ 47 ] In Goodis , the Children’s Lawyer urged the court to refuse or limit the standing of the Information and Privacy Commissioner, whose decision was under review. The Ontario Court of Appeal declined to apply any formal, fixed rule that would limit the tribunal to certain categories of submissions and instead adopted a contextual, discretionary approach: Goodis , at paras. 32-34 . The court found no principled basis for the categorical approach, and observed that such an approach may lead to undesirable consequences:

For example, a categorical rule denying standing if the attack asserts a denial of natural justice could deprive the court of vital submissions if the attack is based on alleged deficiencies in the structure or operation of the tribunal, since these are submissions that the tribunal is uniquely placed to make. Similarly, a rule that would permit a tribunal standing to defend its decision against the standard of reasonableness but not against one of correctness, would allow unnecessary and prevent useful argument. Because the best argument that a decision is reasonable may be that it is correct, a rule based on this distinction seems tenuously founded at best as Robertson J.A. said in United Brotherhood of Carpenters and Joiners of America, Local 1386 v. Bransen Construction Ltd. , 2002 NBCA 27 , [2002] N.B.J. No. 114, 249 N.B.R. (2d) 93 (C.A.), at para. 32 .

( Goodis , at para. 34 )

[ 48 ] The court held that Northwestern Utilities and Paccar should be read as the source of “fundamental considerations” that should guide the court’s exercise of discretion in the context of the case: Goodis , at para. 35 . The two most important considerations, drawn from those cases, were the “importance of having a fully informed adjudication of the issues before the court” (para. 37), and “the importance of maintaining tribunal impartiality”: para. 38. The court should limit tribunal participation if it will undermine future confidence in its objectivity. The court identified a list of factors, discussed further below, that may aid in determining whether and to what extent the tribunal should be permitted to make submissions: paras. 36-38.

[ 49 ] In Canada (Attorney General) v. Quadrini , 2010 FCA 246 , [2012] 2 F.C.R. 3, Stratas J.A. identified two common law restrictions that, in his view, restricted the scope of a tribunal’s participation on appeal from its own decision: finality and impartiality. Finality, the principle whereby a tribunal may not speak on a matter again once it has decided upon it and provided reasons for its decision, is discussed in greater detail below, as it is more directly related to concerns surrounding “bootstrapping” rather than agency standing itself.

[ 50 ] The principle of impartiality is implicated by tribunal argument on appeal, because decisions may in some cases be remitted to the tribunal for further consideration. Stratas J.A. found that “[s]ubmissions by the tribunal in a judicial review proceeding that descend too far, too intensely, or too aggressively into the merits of the matter before the tribunal may disable the tribunal from conducting an impartial redetermination of the merits later”: Quadrini , at para. 16 . However, he ultimately found that these principles did not mandate “hard and fast rules”, and endorsed the discretionary approach set out by the Ontario Court of Appeal in Goodis : Quadrini , at paras. 19-20 .

[ 51 ] A third example of recent judicial consideration of this issue may be found in Leon’s Furniture Ltd. v. Information and Privacy Commissioner (Alta.) , 2011 ABCA 94 , 502 A.R. 110. In this case, Leon’s Furniture challenged the Commissioner’s standing to make submissions on the merits of the appeal (para. 16). The Alberta Court of Appeal, too, adopted the position that the law should respond to the fundamental concerns raised in Northwestern Utilities but should nonetheless approach the question of tribunal standing with discretion, to be exercised in view of relevant contextual considerations: paras. 28-29.

[ 52 ] The considerations set forth by this Court in Northwestern Utilities reflect fundamental concerns with regard to tribunal participation on appeal from the tribunal’s own decision. However, these concerns should not be read to establish a categorical ban on tribunal participation on appeal. A discretionary approach, as discussed by the courts in Goodis , Leon’s Furniture , and Quadrini , provides the best means of ensuring that the principles of finality and impartiality are respected without sacrificing the ability of reviewing courts to hear useful and important information and analysis: see N. Semple, “The Case for Tribunal Standing in Canada” (2007), 20 C.J.A.L.P. 305; L. A. Jacobs and T. S. Kuttner, “Discovering What Tribunals Do: Tribunal Standing Before the Courts” (2002), 81 Can. Bar Rev. 616; F. A. V. Falzon, “Tribunal Standing on Judicial Review” (2008), 21 C.J.A.L.P. 21.

[ 53 ] Several considerations argue in favour of a discretionary approach. Notably, because of their expertise and familiarity with the relevant administrative scheme, tribunals may in many cases be well positioned to help the reviewing court reach a just outcome. For example, a tribunal may be able to explain how one interpretation of a statutory provision might impact other provisions within the regulatory scheme, or the factual and legal realities of the specialized field in which they work. Submissions of this type may be harder for other parties to present.

[ 54 ] Some cases may arise in which there is simply no other party to stand in opposition to the party challenging the tribunal decision. Our judicial review processes are designed to function best when both sides of a dispute are argued vigorously before the reviewing court. In a situation where no other well-informed party stands opposed, the presence of a tribunal as an adversarial party may help the court ensure it has heard the best of both sides of a dispute.

[ 55 ] Canadian tribunals occupy many different roles in the various contexts in which they operate. This variation means that concerns regarding tribunal partiality may be more or less salient depending on the case at issue and the tribunal’s structure and statutory mandate. As such, statutory provisions addressing the structure, processes and role of the particular tribunal are key aspects of the analysis.

[ 56 ] The mandate of the Board, and similarly situated regulatory tribunals, sets them apart from those tribunals whose function it is to adjudicate individual conflicts between two or more parties. For tribunals tasked with this latter responsibility, “the importance of fairness, real and perceived, weighs more heavily” against tribunal standing: Henthorne v. British Columbia Ferry Services Inc. , 2011 BCCA 476 , 344 D.L.R. (4th) 292, at para. 42 .

[ 57 ] I am thus of the opinion that tribunal standing is a matter to be determined by the court conducting the first-instance review in accordance with the principled exercise of that court’s discretion. In exercising its discretion, the court is required to balance the need for fully informed adjudication against the importance of maintaining tribunal impartiality.

[ 58 ] In this case, as an initial matter, the Ontario Energy Board Act, 1998 expressly provides that “[t]he Board is entitled to be heard by counsel upon the argument of an appeal” to the Divisional Court: s. 33(3) . This provision neither expressly grants the Board standing to argue the merits of the decision on appeal, nor does it expressly limit the Board to jurisdictional or standard-of-review arguments as was the case for the relevant statutory provision in Quadrini : see para. 2 .

[ 59 ] In accordance with the foregoing discussion of tribunal standing, where the statute does not clearly resolve the issue, the reviewing court must rely on its discretion to define the tribunal’s role on appeal. While not exhaustive, I would find the following factors, identified by the courts and academic commentators cited above, are relevant in informing the court’s exercise of this discretion:

(1) If an appeal or review were to be otherwise unopposed, a reviewing court may benefit by exercising its discretion to grant tribunal standing.

(2) If there are other parties available to oppose an appeal or review, and those parties have the necessary knowledge and expertise to fully make and respond to arguments on appeal or review, tribunal standing may be less important in ensuring just outcomes.

(3) Whether the tribunal adjudicates individual conflicts between two adversarial parties, or whether it instead serves a policy-making, regulatory or investigative role, or acts on behalf of the public interest, bears on the degree to which impartiality concerns are raised. Such concerns may weigh more heavily where the tribunal served an adjudicatory function in the proceeding that is the subject of the appeal, while a proceeding in which the tribunal adopts a more regulatory role may not raise such concerns.

[ 60 ] Consideration of these factors in the context of this case leads me to conclude that it was not improper for the Board to participate in arguing in favour of the reasonableness of its decision on appeal. First, the Board was the only respondent in the initial review of its decision. Thus, it had no alternative but to step in if the decision was to be defended on the merits. Unlike some other provinces, Ontario has no designated utility consumer advocate, which left the Board — tasked by statute with acting to safeguard the public interest — with few alternatives but to participate as a party.

[ 61 ] Second, the Board is tasked with regulating the activities of utilities, including those in the electricity market. Its regulatory mandate is broad. Among its many roles: it licenses market participants, approves the development of new transmission and distribution facilities, and authorizes rates to be charged to consumers. In this case, the Board was exercising a regulatory role by setting just and reasonable payment amounts to a utility. This is unlike situations in which a tribunal may adjudicate disputes between two parties, in which case the interests of impartiality may weigh more heavily against full party standing.

[ 62 ] The nature of utilities regulation further argues in favour of full party status for the Board here, as concerns about the appearance of partiality are muted in this context. As noted by Doherty J.A., “[l]ike all regulated bodies, I am sure Enbridge wins some and loses some before the [Board]. I am confident that Enbridge fully understands the role of the regulator and appreciates that each application is decided on its own merits by the [Board]”: Enbridge , at para. 28. Accordingly, I do not find that the Board’s participation in the instant appeal was improper. It remains to consider whether the content of the Board’s arguments was appropriate.

(2) Bootstrapping

[ 63 ] The issue of tribunal “bootstrapping” is closely related to the question of when it is proper for a tribunal to act as a party on appeal or judicial review of its decision. The standing issue concerns what types of argument a tribunal may make, i.e. jurisdictional or merits arguments, while the bootstrapping issue concerns the content of those arguments.

[ 64 ] As the term has been understood by the courts who have considered it in the context of tribunal standing, a tribunal engages in bootstrapping where it seeks to supplement what would otherwise be a deficient decision with new arguments on appeal: see, e.g., United Brotherhood of Carpenters and Joiners of America, Local 1386 v. Bransen Construction Ltd. , 2002 NBCA 27 , 249 N.B.R. (2d) 93. Put differently, it has been stated that a tribunal may not “defen[d] its decision on a ground that it did not rely on in the decision under review”: Goodis , at para. 42 .

[ 65 ] The principle of finality dictates that once a tribunal has decided the issues before it and provided reasons for its decision, “absent a power to vary its decision or rehear the matter, it has spoken finally on the matter and its job is done”: Quadrini , at para. 16 , citing Chandler v. Alberta Association of Architects , 1989 41 (SCC) , [1989] 2 S.C.R. 848. Under this principle, the court found that tribunals could not use judicial review as a chance to “amend, vary, qualify or supplement its reasons”: Quadrini , at para. 16 . In Leon’s Furniture , Slatter J.A. reasoned that a tribunal could “offer interpretations of its reasons or conclusion, [but] cannot attempt to reconfigure those reasons, add arguments not previously given, or make submissions about matters of fact not already engaged by the record”: para. 29.

[ 66 ] By contrast, in Goodis , Goudge J.A. found on behalf of a unanimous court that while the Commissioner had relied on an argument not expressly set out in her original decision, this argument was available for the Commissioner to make on appeal. Though he recognized that “[t]he importance of reasoned decision making may be undermined if, when attacked in court, a tribunal can simply offer different, better, or even contrary reasons to support its decision” (para. 42), Goudge J.A. ultimately found that the Commissioner was permitted to raise a new argument on judicial review. The new argument presented was “not inconsistent with the reason offered in the decision. Indeed it could be said to be implicit in it”: para. 55. “It was therefore proper for the Commissioner to be permitted to raise this argument before the Divisional Court and equally proper for the court to decide on that basis”: para. 58.

[ 67 ] There is merit in both positions on the issue of bootstrapping. On the one hand, a permissive stance toward new arguments by tribunals on appeal serves the interests of justice insofar as it ensures that a reviewing court is presented with the strongest arguments in favour of both sides: Semple, at p. 315. This remains true even if those arguments were not included in the tribunal’s original reasons. On the other hand, to permit bootstrapping may undermine the importance of reasoned, well-written original decisions. There is also the possibility that a tribunal, surprising the parties with new arguments in an appeal or judicial review after its initial decision, may lead the parties to see the process as unfair. This may be particularly true where a tribunal is tasked with adjudicating matters between two private litigants, as the introduction of new arguments by the tribunal on appeal may give the appearance that it is “ganging up” on one party. As discussed, however, it may be less appropriate in general for a tribunal sitting in this type of role to participate as a party on appeal.

[ 68 ] I am not persuaded that the introduction of arguments by a tribunal on appeal that interpret or were implicit but not expressly articulated in its original decision offends the principle of finality. Similarly, it does not offend finality to permit a tribunal to explain its established policies and practices to the reviewing court, even if those were not described in the reasons under review. Tribunals need not repeat explanations of such practices in every decision merely to guard against charges of bootstrapping should they be called upon to explain them on appeal or review. A tribunal may also respond to arguments raised by a counterparty. A tribunal raising arguments of these types on review of its decision does so in order to uphold the initial decision; it is not reopening the case and issuing a new or modified decision. The result of the original decision remains the same even if a tribunal seeks to uphold that effect by providing an interpretation of it or on grounds implicit in the original decision.

[ 69 ] I am not, however, of the opinion that tribunals should have the unfettered ability to raise entirely new arguments on judicial review. To do so may raise concerns about the appearance of unfairness and the need for tribunal decisions to be well reasoned in the first instance. I would find that the proper balancing of these interests against the reviewing courts’ interests in hearing the strongest possible arguments in favour of each side of a dispute is struck when tribunals do retain the ability to offer interpretations of their reasons or conclusions and to make arguments implicit within their original reasons: see Leon’s Furniture , at para. 29; Goodis , at para. 55 .

[ 70 ] In this case, I do not find that the Board impermissibly stepped beyond the bounds of its original decision in its arguments before this Court. In its reply factum, the Board pointed out — correctly, in my view — that its submissions before this Court simply highlight what is apparent on the face of the record, or respond to arguments raised by the respondents.

[ 71 ] I would, however, urge the Board, and tribunal parties in general, to be cognizant of the tone they adopt on review of their decisions. As Goudge J.A. noted in Goodis :

. . . if an administrative tribunal seeks to make submissions on a judicial review of its decision, it [should] pay careful attention to the tone with which it does so. Although this is not a discrete basis upon which its standing might be limited, there is no doubt that the tone of the proposed submissions provides the background for the determination of that issue. A tribunal that seeks to resist a judicial review application will be of assistance to the court to the degree its submissions are characterized by the helpful elucidation of the issues, informed by its specialized position, rather than by the aggressive partisanship of an adversary. [para. 61]

[ 72 ] In this case, the Board generally acted in such a way as to present helpful argument in an adversarial but respectful manner. However, I would sound a note of caution about the Board’s assertion that the imposition of the prudent investment test “would in all likelihood not change the result” if the decision were remitted for reconsideration (A.F., at para. 99). This type of statement may, if carried too far, raise concerns about the principle of impartiality such that a court would be justified in exercising its discretion to limit tribunal standing so as to safeguard this principle.

B. Standard of Review

[ 73 ] The parties do not dispute that reasonableness is the appropriate standard of review for the Board’s actions in applying its expertise to set rates and approve payment amounts under the Ontario Energy Board Act, 1998 . I agree. In addition, to the extent that the resolution of this appeal turns on the interpretation of the Ontario Energy Board Act, 1998 , the Board’s home statute, a standard of reasonableness presumptively applies: Dunsmuir v. New Brunswick , 2008 SCC 9 , [2008] 1 S.C.R. 190, at para. 54 ; Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association , 2011 SCC 61 , [2011] 3 S.C.R. 654, at para. 30 ; Tervita Corp. v. Canada (Commissioner of Competition) , 2015 SCC 3 , [2015] 1 S.C.R. 161, at para. 35 . Nothing in this case suggests the presumption should be rebutted.

[ 74 ] This appeal involves two distinct uses of the term “reasonable”. One concerns the standard of review: on appeal, this Court is charged with evaluating the “justification, transparency and intelligibility” of the Board’s reasoning, and “whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law” ( Dunsmuir , at para. 47 ). The other is statutory: the Board’s rate-setting powers are to be used to ensure that, in its view, a just and reasonable balance is struck between utility and consumer interests. These reasons will attempt to keep the two uses of the term distinct.

C. Choice of Methodology Under the Ontario Energy Board Act, 1998

[ 75 ] The question of whether the Board’s decision to disallow recovery of certain costs was reasonable turns on how that decision relates to the Board’s statutory and regulatory powers to approve payments to utilities and to have these payments reflected in the rates paid by consumers. The Board’s general rate- and payment-setting powers are described above under the “Regulatory Framework” heading.

[ 76 ] The just-and-reasonable approach to recovery of the cost of services provided by a utility captures the essential balance at the heart of utilities regulation: to encourage investment in a robust utility infrastructure and to protect consumer interests, utilities must be allowed, over the long run, to earn their cost of capital, no more, no less.

[ 77 ] The Ontario Energy Board Act, 1998 does not, however, either in s. 78.1 or elsewhere, prescribe the methodology the Board must use to weigh utility and consumer interests when deciding what constitutes just and reasonable payment amounts to the utility. Indeed, s. 6(1) of O. Reg. 53/05 expressly permits the Board, subject to certain exceptions set out in s. 6(2), to “establish the form, methodology , assumptions and calculations used in making an order that determines payment amounts for the purpose of section 78.1 of the Act”.

[ 78 ] As a contrasting example, para. 4.1 of s. 6(2) of O. Reg. 53/05 establishes a specific methodology for use when the Board reviews “costs incurred and firm financial commitments made in the course of planning and preparation for the development of proposed new nuclear generation facilities”. When reviewing such costs, the Board must be satisfied that “the costs were prudently incurred ” and that “the financial commitments were prudently made ”: para. 4.1 of 6(2). The provision thus establishes a specific context in which the Board’s analysis is focused on the prudence of the decision to incur or commit to certain costs. The absence of such language in the more general s. 6(1) provides further reason to read the regulation as providing broad methodological discretion to the Board in making orders for payment amounts where the specific provisions of s. 6(2) do not apply.

[ 79 ] Regarding whether a presumption of prudence must be applied to OPG’s decisions to incur costs, neither the Ontario Energy Board Act, 1998 nor O. Reg. 53/05 expressly establishes such a presumption. Indeed, the Ontario Energy Board Act, 1998 places the burden on the applicant utility to establish that payment amounts approved by the Board are just and reasonable: s. 78.1(6) and (7). It would thus seem inconsistent with the statutory scheme to presume that utility decisions to incur costs were prudent.

[ 80 ] Justice Abella concludes that the Board’s review of OPG’s costs should have consisted of “an after-the-fact prudence review, with a rebuttable presumption that the utility’s expenditures were reasonable”: para. 150. Such an approach is contrary to the statutory scheme. While the Board has considerable methodological discretion, it does not have the freedom to displace the burden of proof established by s. 78.1(6) of the Ontario Energy Board Act, 1998 : “. . . the burden of proof is on the applicant in an application made under this section”. Of course, this does not imply that the applicant must systematically prove that every single cost is just and reasonable. The Board has broad discretion to determine the methods it may use to examine costs — it just cannot shift the burden of proof contrary to the statutory scheme.

[ 81 ] In judicially reviewing a decision of the Board to allow or disallow payments to a utility, the court’s role is to assess whether the Board reasonably determined that a certain payment amount was “just and reasonable” for both the utility and the consumers. Such an approach is consistent with this Court’s rate-setting jurisprudence in other regulatory domains in which the regulator is given methodological discretion, where it has been observed that “[t]he obligation to act is a question of law, but the choice of the method to be adopted is a question of discretion with which, under the statute, no Court of law may interfere”: Bell Canada v. Bell Aliant Regional Communications , 2009 SCC 40 , [2009] 2 S.C.R. 764, at para. 40 (concerning telecommunication rate-setting), quoting Re General Increase in Freight Rates (1954), 76 C.R.T.C. 12 (S.C.C.), at p. 13 (concerning railway freight rates) . Of course, today this statement must be understood to permit intervention by a court where the exercise of discretion rendered a decision unreasonable. Accordingly, it remains to determine whether the Board’s analytical approach to disallowing the costs at issue in this case rendered the Board’s decision unreasonable under the “just and reasonable” standard.

D. Characterization of Costs at Issue

[ 82 ] Forecast costs are costs which the utility has not yet paid, and over which the utility still retains discretion as to whether the disbursement will be made. A disallowance of such costs presents a utility with a choice: it may change its plans and avoid the disallowed costs, or it may incur the costs regardless of the disallowance with the knowledge that the costs will ultimately be borne by the utility’s shareholders rather than its ratepayers. By contrast, committed costs are those for which, if a regulatory board disallows recovery of the costs in approved payments, the utility and its shareholders will have no choice but to bear the burden of those costs themselves. This result may occur because the utility has already spent the funds, or because the utility entered into a binding commitment or was subject to other legal obligations that leave it with no discretion as to whether to make the payment in the future.

[ 83 ] There is disagreement between the parties as to how the costs disallowed by the Board in this matter should be characterized. The Board asserts that compensation costs for the test period are forecast insofar as they have not yet been disbursed, while OPG asserts that the costs should be characterized as committed, because OPG is under a contractual obligation to pay those amounts when they become due. This disagreement is important because a “no hind-sight” prudence review, which is discussed in detail below, has developed in the context of “committed” costs. Indeed, it makes no sense to apply such a test where a utility still retains discretion over whether the costs will ultimately be incurred; the decision to commit the utility to such costs has not yet been made. Accordingly, where the regulator has discretion over its methodological approach, understanding whether the costs at issue are “forecast” or “committed” may be helpful in reviewing the reasonableness of a regulator’s choice of methodology.

[ 84 ] In this case, at least some of the compensation costs that the Board found to be excessive were driven by collective agreements to which OPG had committed before the application at issue, and which established compensation costs that were, in aggregate, above the 75th percentile for comparable positions at other utilities. The collective agreements left OPG with limited flexibility regarding overall compensation rates or staffing levels — OPG was required to abide by wage and staffing levels established by collective agreements, and retained flexibility only over terms outside the bounds of those agreements — and thus those portions of OPG’s compensation rates and staffing levels that were dictated by the terms of the collective agreements were committed costs.

[ 85 ] However, the Board found that OPG’s compensation costs for the test period were not entirely driven by the collective agreements, and thus were not entirely committed, because OPG retained some flexibility to manage total staffing levels in light of projected attrition of a mature workforce. The Board Decision did not, however, include detailed forecasts regarding exactly how much of the $145 million in disallowed compensation costs could be recovered through natural reduction in employee numbers or other adjustments, and how much would necessarily be borne by the utility and its shareholder. Accordingly, the disallowed costs at issue must be understood as being at least partially committed. It is unreasonable to characterize them as entirely forecast in view of the constraints placed on OPG by the collective agreements.

[ 86 ] Having established that the disallowed costs are at least partially committed, it is necessary to consider whether the Board acted reasonably in not applying a no-hindsight prudent investment test in assessing those costs. Accordingly, I now turn to the jurisprudential history and methodological details of the prudent investment test.

E. The Prudent Investment Test

[ 87 ] In order to assess whether the Board’s methodology was reasonable in this case, it is necessary to provide some background on the prudent investment test (sometimes referred to as “prudence review” or the “prudence test”) in order to identify its origins, place it in context, and explore how it has been understood by utilities, regulators, and legislators.

(1) American Jurisprudence

[ 88 ] American jurisprudence has played a significant role in the history of the prudent investment test in utilities regulation. In discussing this history, I would first reiterate this Court’s observation that “[w]hile the American jurisprudence and texts in this area should be considered with caution given that Canada and the United States have very different political and constitutional-legal regimes, they do shed some light on the issue”: ATCO Gas and Pipelines Ltd. v. Alberta (Energy and Utilities Board) , 2006 SCC 4 , [2006] 1 S.C.R. 140, at para. 54 .

[ 89 ] The origins of the prudent investment test in the context of utilities regulation may be traced to Justice Brandeis of the Supreme Court of the United States, who wrote a concurring opinion in 1923 to observe that utilities should receive deference in seeking to recover “investments which, under ordinary circumstances, would be deemed reasonable”: State of Missouri ex rel. Southwestern Bell Telephone Co. v. Public Service Commission of Missouri , 262 U.S. 276 (1923), at p. 289, fn.1.

[ 90 ] In the decades that followed, American utility regulators tasked with reviewing past-incurred utility costs generally employed one of two standards: the “used and useful” test or the “prudent investment” test (J. Kahn, “Keep Hope Alive: Updating the Prudent Investment Standard for Allocating Nuclear Plant Cancellation Costs” (2010), 22 Fordham Envtl. L. Rev. 43, at p. 49). These tests took different approaches to determining what costs could justly and reasonably be passed on to ratepayers. The used and useful test allowed utilities to earn returns only on those investments that were actually used and useful to the utility’s operations, on the principle that ratepayers should not be compelled to pay for investments that do not benefit them.

[ 91 ] By contrast, the prudent investment test followed Justice Brandeis’s preferred approach by allowing for recovery of costs provided they were not imprudent based on what was known at the time the investment or expense was incurred: Kahn, at pp. 49-50. Though it may seem problematic from the perspective of consumer interests to adopt the prudent investment test — a test that allows for payments related to investments that may not be used or useful — it gives regulators a tool to soften the potentially harsh effects of the used and useful test, which may place onerous burdens on utilities. Disallowing recovery of the cost of failed investments that appeared reasonable at the time, for example, may imperil the financial health of utilities, and may chill the incentive to make such investments in the first place. This effect may then have negative implications for consumers, whose long-run interests will be best served by a dynamically efficient and viable electricity industry. Thus, the prudent investment test may be employed by regulators to strike the appropriate balance between consumer and utility interests: see Kahn, at pp. 53-54.

[ 92 ] The states differed in their approaches to setting the statutory foundation for utility regulation. Regulators in some states were free to apply the prudent investment test, while other states enacted statutory provisions disallowing compensation in respect of capital investments that were not “used and useful in service to the public”: Duquesne Light Co. v. Barasch , 488 U.S. 299 (1989), at p. 302. Notably, when asked in Duquesne to consider whether “just and reasonable” payments to utilities required, as a constitutional matter, that the prudent investment test be applied to past-incurred costs, the U.S. Supreme Court held that “[t]he designation of a single theory of ratemaking as a constitutional requirement would unnecessarily foreclose alternatives which could benefit both consumers and investors”: p. 316.

[ 93 ] American courts have also recognized that there may exist some contexts in which certain features of the prudent investment test may be less justifiable. For example, the Supreme Court of Utah considered whether a presumption of reasonableness was justified when reviewing costs passed to a utility by an unregulated affiliate entity, and concluded that it was not appropriate:

. . . we do not think an affiliate expense should carry a presumption of reasonableness. While the pressures of a competitive market might allow us to assume, in the absence of a showing to the contrary, that nonaffiliate expenses are reasonable, the same cannot be said of affiliate expenses not incurred in an arm’s length transaction.

( U.S. West Communications, Inc. v. Public Service Commission of Utah , 901 P.2d 270 (Utah 1995), at p. 274)

[ 94 ] Treatment of the prudent investment test in American jurisprudence thus indicates that the test has been employed as a tool that may be useful in arriving at just and reasonable outcomes, rather than a mandatory feature of utilities regulation that must be applied regardless of whether there is statutory language to that effect.

(2) Canadian Jurisprudence

[ 95 ] Following its emergence in American jurisprudence, several Canadian utility regulators and courts have also considered the role of prudence review and, in some cases, applied a form of the prudent investment test. I provide a review of some of these cases here not in an attempt to exhaustively catalogue all uses of the test, but rather to set out the way in which the test has been invoked in various contexts.

[ 96 ] In British Columbia Electric Railway Co. v. Public Utilities Commission of British Columbia , 1960 44 (SCC) , [1960] S.C.R. 837, Martland J. observed that the statute at issue in that case directed that the regulator, in fixing rates,

( a ) . . shall consider all matters which it deems proper as affecting the rate: [and]

( b ) . . . shall have due regard, among other things, to the protection of the public from rates that are excessive as being more than a fair and reasonable charge for services of the nature and quality furnished by the public utility; and to giving to the public utility a fair and reasonable return upon the appraised value of the property of the public utility used, or prudently and reasonably acquired, to enable the public utility to furnish the service. [p. 852]

(Quoting Public Utilities Act , R.S.B.C. 1948, c. 277, s. 16(1)( b ) (repealed S.B.C. 1973, c. 29, s. 187).)

The consequence of this statutory language, Martland J. held, was that the regulator, “when dealing with a rate case, has unlimited discretion as to the matters which it may consider as affecting the rate, but that it must, when actually setting the rate, meet the two requirements specifically mentioned in clause ( b )”: p. 856. That is, the regulator, under this statute, must ensure that the public pays only fair and reasonable charges, and that the utility secures a fair and reasonable return upon its property used or prudently and reasonably acquired . This express statutory protection for the recovery of prudently made property acquisition costs thus provides an example of statutory language under which this Court found a non-discretionary obligation to provide a fair return to utilities for capital expenditures that were either used or prudently acquired.

[ 97 ] In 2005, the Nova Scotia Utility and Review Board (“NSUARB”) considered and adopted a definition of the prudent investment test articulated by the Illinois Commerce Commission:

. . . prudence is that standard of care which a reasonable person would be expected to exercise under the same circumstances encountered by utility management at the time decisions had to be made. . . . Hindsight is not applied in assessing prudence. . . . A utility’s decision is prudent if it was within the range of decisions reasonable persons might have made. . . . The prudence standard recognizes that reasonable persons can have honest differences of opinion without one or the other necessarily being imprudent.

( Nova Scotia Power Inc., Re , 2005 NSUARB 27 (“ Nova Scotia Power 2005 ”), at para. 84)

The NSUARB then wrote that “[f]ollowing a review of the cases, the Board finds that the definition of imprudence as set out by the Illinois Commerce Commission is a reasonable test to be applied in Nova Scotia”: para. 90. The NSUARB then considered, among other things, whether the utility’s recent fuel procurement strategy had been prudent, and found that it had not: para. 94. It did not, however, indicate that it believed itself to be compelled to apply the prudent investment test.

[ 98 ] The NSUARB reaffirmed its endorsement of the prudent investment test in 2012: Nova Scotia Power Inc. (Re) , 2012 NSUARB 227 (“ Nova Scotia Power 2012 ”), at paras. 143-46. In that case, the utility whose submissions were under review “confirmed that from its perspective this is the test the Board should apply”: para. 146. The NSUARB then applied the prudence test in evaluating whether several of the utility’s operational decisions were prudent, and found that some were not: para. 188.

[ 99 ] In 2006, the Ontario Court of Appeal considered the meaning of the prudent investment test in Enbridge . This case is of particular interest for two reasons. First, the Ontario Court of Appeal endorsed in its reasons a specific formulation of the prudent investment test framework:

– Decisions made by the utility’s management should generally be presumed to be prudent unless challenged on reasonable grounds.

– To be prudent, a decision must have been reasonable under the circumstances that were known or ought to have been known to the utility at the time the decision was made.

– Hindsight should not be used in determining prudence, although consideration of the outcome of the decision may legitimately be used to overcome the presumption of prudence.

– Prudence must be determined in a retrospective factual inquiry, in that the evidence must be concerned with the time the decision was made and must be based on facts about the elements that could or did enter into the decision at the time. [para. 10]

[ 100 ] Second, the Court of Appeal in Enbridge made certain statements that suggest that the prudent investment test was a necessary approach to reviewing committed costs. Specifically, it noted that in deciding whether Enbridge’s requested rate increase was just and reasonable,

the [Board] was required to balance the competing interests of Enbridge and its consumers. That balancing process is achieved by the application of what is known in the utility rate regulation field as the “prudence” test. Enbridge was entitled to recover its costs by way of a rate increase only if those costs were “prudently” incurred. [para. 8]

The Court of Appeal also noted that the Board had applied the “proper test”: para. 18. These statements tend to suggest that the Court of Appeal was of the opinion that prudence review is an inherent and necessary part of ensuring just and reasonable payments.

[ 101 ] However, the question of whether the prudence test was a required feature of just-and-reasonable analysis in this context was not squarely before the Court of Appeal in Enbridge . Rather, the parties in that case “were in substantial agreement on the general approach the Board should take to reviewing the prudence of a utility’s decision” (para. 10), and the question at issue was whether the Board had reasonably applied that agreed-upon approach. In this sense, Enbridge is similar to Nova Scotia Power 2012 : both cases involved the application of prudence analysis in contexts where there was no dispute over whether an alternative methodology could reasonably have been applied.

(3) Conclusion Regarding the Prudent Investment Test

[ 102 ] The prudent investment test, or prudence review, is a valid and widely accepted tool that regulators may use when assessing whether payments to a utility would be just and reasonable. While there exist different articulations of prudence review, Enbridge presents one express statement of how a regulatory board might structure its review to assess the prudence of utility expenditures at the time they were incurred or committed. A no-hindsight prudence review has most frequently been applied in the context of capital costs, but Enbridge and Nova Scotia Power (both 2005 and 2012 ) provide examples of its application to decisions regarding operating costs as well. I see no reason in principle why a regulatory board should be barred from applying the prudence test to operating costs.

[ 103 ] However, I do not find support in the statutory scheme or the relevant jurisprudence for the notion that the Board should be required as a matter of law, under the Ontario Energy Board Act, 1998 , to apply the prudence test as outlined in Enbridge such that the mere decision not to apply it when considering committed costs would render its decision on payment amounts unreasonable. Nor is the creation of such an obligation by this Court justified. As discussed above, where a statute requires only that the regulator set “just and reasonable” payments, as the Ontario Energy Board Act, 1998 does in Ontario, the regulator may make use of a variety of analytical tools in assessing the justness and reasonableness of a utility’s proposed payment amounts. This is particularly so where, as here, the regulator has been given express discretion over the methodology to be used in setting payment amounts: O. Reg. 53/05, s. 6(1) .

[ 104 ] To summarize, it is not necessarily unreasonable, in light of the particular regulatory structure established by the Ontario Energy Board Act, 1998 , for the Board to evaluate committed costs using a method other than a no-hindsight prudence review. As noted above, applying a presumption of prudence would have conflicted with the burden of proof in the Ontario Energy Board Act, 1998 and would therefore not have been reasonable. The question of whether it was reasonable to assess a particular cost using hindsight should turn instead on the circumstances of that cost. I emphasize, however, that this decision should not be read to give regulators carte blanche to disallow a utility’s committed costs at will. Prudence review of committed costs may in many cases be a sound way of ensuring that utilities are treated fairly and remain able to secure required levels of investment capital. As will be explained, particularly with regard to committed capital costs, prudence review will often provide a reasonable means of striking the balance of fairness between consumers and utilities.

[ 105 ] This conclusion regarding the Board’s ability to select its methodology rests on the particulars of the statutory scheme under which the Board operates. There exist other statutory schemes in which regulators are expressly required to compensate utilities for certain costs prudently incurred: see British Columbia Electric Railway Co . Under such a framework, the regulator’s methodological discretion may be more constrained.

(4) Application to the Board’s Decision

[ 106 ] In this case, the Board disallowed a total of $145 million in compensation costs associated with OPG’s nuclear operations, over two years. As discussed above, these costs are best understood as at least partly committed. In view of the nature of these particular costs and the circumstances in which they became committed, I do not find that the Board acted unreasonably in not applying the prudent investment test in determining whether it would be just and reasonable to compensate OPG for these costs.

[ 107 ] First, the costs at issue are operating costs, rather than capital costs. Capital costs, particularly those pertaining to areas such as capacity expansion or upgrades to existing facilities, often entail some amount of risk, and may not always be strictly necessary to the short-term ongoing production of the utility. Nevertheless, such costs may often be a wise investment in the utility’s future health and viability. As such, prudence review, including a no-hindsight approach (with or without a presumption of prudence, depending on the applicable statutory context), may play a particularly important role in ensuring that utilities are not discouraged from making the optimal level of investment in the development of their facilities.

[ 108 ] Operating costs, like those at issue here, are different in kind from capital costs. There is little danger in this case that a disallowance of these costs will have a chilling effect on OPG’s willingness to incur operating costs in the future, because costs of the type disallowed here are an inescapable element of operating a utility. It is true that a decision such as the Board’s in this case may have the effect of making OPG more hesitant about committing to relatively high compensation costs, but that was precisely the intended effect of the Board’s decision.

[ 109 ] Second, the costs at issue arise in the context of an ongoing, “repeat-player” relationship between OPG and its employees. Prudence review has its origins in the examination of decisions to pursue particular investments, such as a decision to invest in capacity expansion; these are often one-time decisions made in view of a particular set of circumstances known or assumed at the time the decision was made.

[ 110 ] By contrast, OPG’s committed compensation costs arise in the context of an ongoing relationship in which OPG will have to negotiate compensation costs with the same parties in the future. Such a context supports the reasonableness of a regulator’s decision to weigh all evidence it finds relevant in striking a just and reasonable balance between the utility and consumers, rather than confining itself to a no-hindsight approach. Prudence review is simply less relevant when the Board’s focus is not solely on compensating for past commitments, but on regulating costs to be incurred in the future as well. As will be discussed further, the Board’s ultimate disallowance was not targeted exclusively at committed costs, but rather was made with respect to the total compensation costs it evaluated in aggregate. Though the Board acknowledged that OPG may not have had the discretion to reduce spending by the entire amount of the disallowance, the disallowance was animated by the Board’s efforts to get OPG’s ongoing compensation costs under control.

[ 111 ] Having already given OPG a warning that the Board found its operational costs to be of concern (see Board 2008-2009 Decision, at pp. 28-32), it was not unreasonable for the Board to be more forceful in considering compensation costs to ensure effective regulation of such costs going forward. The Board’s statement that its disallowance was intended “to send a clear signal that OPG must take responsibility for improving its performance” (Board Decision, at para. 350) shows that it had the ongoing effects of its disallowance squarely in mind in issuing its decision in this case.

[ 112 ] The reasonableness of the Board’s decision to disallow $145 million in compensation costs is supported by the Board’s recognition of the fact that OPG was bound to a certain extent by the collective agreements in making staffing decisions and setting compensation rates, and its consideration of this factor in setting the total disallowance: Board Decision, at para. 350. The Board’s methodological flexibility ensures that its decision need not be “all or nothing”. Where appropriate, to the extent that the utility was unable to reduce its costs, the total burden of such costs may be moderated or shared as between the utility’s shareholders and the consumers. The Board’s moderation in this case shows that, in choosing to disallow costs without applying a formal no-hindsight prudence review, it remained mindful of the need to ensure that any disallowance was not unfair to OPG and certainly did not impair the viability of the utility.

[ 113 ] Justice Abella, in her dissent, acknowledges that the Board has the power under prudence review to disallow committed costs in at least some circumstances: para. 152. However, she speculates that any such disallowance could “imperil the assurance of reliable electricity service”: para. 156. A large or indiscriminate disallowance might create such peril, but it is also possible for the Board to do as it did here, and temper its disallowance to recognize the realities facing the utility.

[ 114 ] There is no dispute that collective agreements are “immutable” between employees and the utility. However, if the legislature had intended for costs under collective agreements to also be inevitably imposed on consumers, it would not have seen fit to grant the Board oversight of utility compensation costs. The existence both of collective bargaining for utility employees and of the Board’s power to fix payment amounts covering compensation costs indicates neither regime can trump the other. The Board cannot interfere with the collective agreement by ordering that a utility break its obligations thereunder, but nor can the collective agreement supersede the Board’s duty to ensure a just and reasonable balance between utility and consumer interests.

[ 115 ] Justice Abella says that the Board’s review of committed costs using hindsight evidence appears to contradict statements made earlier in its decision. The Board wrote that it would use all relevant evidence in assessing forecast costs but that it would limit itself to a no-hindsight approach in reviewing costs that OPG could not “take action to reduce”: Board Decision, at para. 75. In my view, these statements can be read as setting out a reasonable approach for analyzing costs that could reliably be fit into forecast or committed categories. However, not all costs are amenable to such clean categorization by the Board in assessing payment amounts for a test period.

[ 116 ] With regard to the compensation costs at issue here, the Board declined to split the total cost disallowance into forecast and committed components in conducting its analysis. As Hoy J. observed, “[g]iven the complexity of OPG’s business, and respecting its management’s autonomy, [the Board] did not try to quantify precisely the amount by which OPG could reduce its forecast compensation costs within the framework of the existing collective bargaining agreements”: Div. Ct. reasons, at para. 53. That is, the Board did not split all compensation costs into either “forecast” or “committed”, but analyzed the disallowance of compensation costs as a mix of forecast and committed expenditures over which management retained some, but not total, control.

[ 117 ] It was not unreasonable for the Board to proceed on the basis that predicting staff attrition rates is an inherently uncertain exercise, and that it is not equipped to micromanage business decisions within the purview of OPG management. These considerations mean that any attempt to predict the exact degree to which OPG would be able to reduce compensation costs (in other words, what share of the costs were forecast) would be fraught with uncertainty. Accordingly, it was not unreasonable for the Board to adopt a mixed approach that did not rely on quantifying the exact share of compensation costs that fell into the forecast and committed categories. Such an approach is not inconsistent with the Board’s discussion at paras. 73-75, but rather represents an exercise of the Board’s methodological discretion in addressing a challenging issue where these costs did not fit easily into the categories discussed in that passage.

[ 118 ] Justice Abella emphasizes throughout her reasons that the costs established by the collective agreements were not adjustable. I do not dispute this point. However, to the extent that she relies on the observation that the collective agreements “made it illegal for the utility to alter the compensation and staffing levels” of the unionized workforce (para. 149 (emphasis in original)), one might conclude that the Board was in some way trying to interfere with OPG’s obligations under its collective agreements. It is important not to lose sight of the fact that the Board decision in no way purports to force OPG to break its contractual commitments to unionized employees.

[ 119 ] Finally, her observation that the Canadian Nuclear Safety Commission (“CNSC”) “has . . . imposed staffing levels on Ontario Power Generation to ensure safe and reliable operation of its nuclear stations” (para. 127) is irrelevant to the issues raised in this case. While the regime put in place by the CNSC surely imposes operational and staffing restraints on nuclear utilities (see OPG record, at pp. 43-46), there is nothing in the Board’s reasons, and no argument presented before this Court, suggesting that the Board’s disallowance will result in a violation of the provisions of the Nuclear Safety and Control Act , S.C. 1997, c. 9 .

[ 120 ] I have noted above that it is essential for a utility to earn its cost of capital in the long run. The Board’s disallowance may have adversely impacted OPG’s ability to earn its cost of capital in the short run. Nevertheless, the disallowance was intended “to send a clear signal that OPG must take responsibility for improving its performance” (Board Decision, at para. 350). Such a signal may, in the short run, provide the necessary impetus for OPG to bring its compensation costs in line with what, in the Board’s opinion, consumers should justly expect to pay for an efficiently provided service. Sending such a signal is consistent with the Board’s market proxy role and its objectives under s. 1 of the Ontario Energy Board Act, 1998 .

VI. Conclusion

[ 121 ] I do not find that the Board acted improperly in pursuing this matter on appeal; nor do I find that it acted unreasonably in disallowing the compensation costs at issue. Accordingly, I would allow the appeal, set aside the decision of the Court of Appeal, and reinstate the decision of the Board.

The following are the reasons delivered by

[ 122 ] Abella J. (dissenting) — The Ontario Energy Board was established in 1960 to set rates for the sale and storage of natural gas and to approve pipeline construction projects. Over time, its powers and responsibilities evolved. In 1973, the Board became responsible for reviewing and reporting to the Minister of Energy on electricity rates. During this period, Ontario’s electricity market was lightly regulated, dominated by the government-owned Ontario Hydro, which owned power generation assets responsible for about 90 per cent of electricity production in the province: Ron W. Clark, Scott A. Stoll and Fred D. Cass, Ontario Energy Law: Electricity (2012), at p. 134; 2011 Annual Report of the Office of the Auditor General of Ontario, at pp. 5 and 67.

[ 123 ] A series of legislative measures in the late 1990s were adopted to transform the electricity industry into a market-based one driven by competition. Ontario Hydro was unbundled into five entities. One of them was Ontario Power Generation Inc., which was given responsibility for controlling the power generation assets of the former Ontario Hydro. It was set up as a commercial corporation with one shareholder — the Province of Ontario: Clark, Stoll and Cass, at pp. 5-7 and 134.

[ 124 ] As of April 1, 2008, the Board was given the authority by statute to set payments for the electricity generated by a prescribed list of assets held by Ontario Power Generation: Ontario Energy Board Act, 1998 , S.O. 1998, c. 15, Sch. B, s. 78.1(2); O. Reg. 53/05 , Payments Under Section 78.1 of the Act , s. 3. Under the legislative scheme, Ontario Power Generation is required to apply to the Board for the approval of “just and reasonable” payment amounts: Ontario Energy Board Act, 1998 , s. 78.1(5). The Board sets its own methodology to determine what “just and reasonable” payment amounts are, guided by the statutory objectives to maintain a “financially viable electricity industry” and to “protect the interests of consumers with respect to prices and the adequacy, reliability and quality of electricity service”: O. Reg. 53/05, s. 6(1) ; Ontario Energy Board Act, 1998 , paras. 1 and 2 of s. 1(1) .

[ 125 ] Ontario Power Generation remains the province’s largest electricity generator. It was unionized by the Ontario Hydro Employees’ Union (the predecessor to the Power Workers’ Union) in the 1950s, and by the Society of Energy Professionals in 1992: Richard P. Chaykowski, An Assessment of the Industrial Relations Context and Outcomes at OPG (2013) (online), at s. 6.2. Today, Ontario Power Generation employs approximately 10,000 people in its regulated businesses, 90 per cent of whom are unionized. Two thirds of these unionized employees are represented by the Power Workers’ Union, and the rest by the Society of Energy Professionals.

[ 126 ] Both the Power Workers’ Union and the Society of Energy Professionals had collective agreements with Ontario Hydro before Ontario Power Generation was established. As a successor company to Ontario Hydro, Ontario Power Generation inherited the full range of these labour relations obligations: Ontario Labour Relations Act, 1995 , S.O. 1995, c. 1, Sch. A, s. 69. Ontario Power Generation’s collective agreements with its unions prevent the utility from unilaterally reducing staffing or compensation levels.

[ 127 ] The Canadian Nuclear Safety Commission, an independent federal government agency responsible for ensuring compliance with the Nuclear Safety and Control Act , S.C. 1997, c. 9, has also imposed staffing levels on Ontario Power Generation to ensure safe and reliable operation of its nuclear stations.

[ 128 ] On May 26, 2010, Ontario Power Generation applied to the Board for a total revenue requirement of $6,909.6 million, including $2,783.9 million in compensation costs — wages, benefits, pension servicing, and annual incentives — to cover the period from January 1, 2011 to December 31, 2012: EB-2010-0008, at pp. 8, 49 and 80.

[ 129 ] In its decision, the Board explained that it would use “two types of examination” to assess the utility’s expenditures. When evaluating forecast costs — costs that the utility has estimated for a future period and which can still be reduced or avoided — the Board said that Ontario Power Generation bears the burden of showing that these costs are reasonable. On the other hand, when the Board would be evaluating costs for which “[t]here is no opportunity for the company to take action to reduce”, otherwise known as committed costs, it said that it would undertake “an after-the-fact prudence review . . . conducted in the manner which includes a presumption of prudence”, that is, a presumption that the utility’s expenditures are reasonable: p. 19.

[ 130 ] The Board made no distinction between those compensation costs that were reducible and those that were not. Instead, it subjected all compensation costs to the kind of assessment it uses for reducible, forecast costs and disallowed $145 million because it concluded that the utility’s compensation rates and staffing levels were too high.

[ 131 ] On appeal, a majority of the Divisional Court upheld the Board’s order. In dissenting reasons, Aitken J. concluded that the Board’s decision was unreasonable because it did not apply the proper approach to the compensation costs which were, as a result of legally binding collective agreements, fixed and not adjustable. Instead, the Board “lumped” all compensation costs together and made no distinction between those that were the result of binding contractual obligations and those that were not. As she said:

First, I consider any limitation on [Ontario Power Generation’s] ability to manage nuclear compensation costs on a go-forward basis, due to binding collective agreements in effect prior to the application and the test period, to be costs previously incurred and subject to an after-the-fact, two-step, prudence review. Second, I conclude that, in considering [Ontario Power Generation’s] nuclear compensation costs, as set out in its application, the [Board] in its analysis (though not necessarily in its final number) was required to differentiate between such earlier incurred liabilities and other aspects of the nuclear compensation cost package that were truly projected and not predetermined. Third, in my view, the [Board] was required to undergo a prudence review in regard to those aspects of the nuclear compensation package that arose under binding contracts entered prior to the application and the test period. In regard to the balance of factors making up the nuclear compensation package, the [Board] was free to determine, based on all available evidence, whether such factors were reasonable. Fourth, had a prudence review been undertaken, there was evidence upon which the [Board] could reasonably have decided that the presumption of prudence had been rebutted in regard to those cost factors mandated in the collective agreements. Unfortunately, I cannot find anywhere in the Decision of the [Board] where such an analysis was undertaken. The [Board] lumped all nuclear compensation costs together. It dealt with them as if they all emanated from the same type of factors and none reflected contractual obligations to which the [Ontario Power Generation] was bound due to a collective agreement entered prior to the application and the test period. Finally, I conclude that, when the [Board] was considering the reasonableness of the nuclear compensation package, it erred in considering evidence that came into existence after the date on which the collective agreements were entered when it assessed the reasonableness of the rates of pay and other binding provisions in the collective agreements. [para. 75]

[ 132 ] The Court of Appeal unanimously agreed with Aitken J.’s conclusion, finding that “the compensation costs at issue before the [Board] were committed costs” which should therefore have been assessed using a presumption of prudence. As they both acknowledged, it was open to the Board to find that the presumption had been rebutted in connection with the binding contractual obligations, but the Board acted unreasonably in failing to take the immutable nature of the fixed costs into consideration.

[ 133 ] I agree. The compensation costs for approximately 90 per cent of Ontario Power Generation’s regulated workforce were established through legally binding collective agreements which obligated the utility to pay fixed levels of compensation, regulated staffing levels, and provided unionized employees with employment security. Ontario Power Generation’s compensation costs were therefore overwhelmingly predetermined and could not be adjusted by the utility during the relevant period. These are precisely the type of costs that the Board referred to in its decision as costs for which “[t]here is no opportunity for the company to take action to reduce” and which must be subjected to “a prudence review conducted in the manner which includes a presumption of prudence”: para. 75.

[ 134 ] In my respectful view, failing to acknowledge the legally binding, non-reducible nature of the cost commitments reflected in the collective agreements and apply the review the Board itself said should apply to such costs, rendered its decision unreasonable.

Analysis

[ 135 ] Pursuant to s. 78.1(5) of the Ontario Energy Board Act, 1998 , upon application from Ontario Power Generation, the Board is required to determine “just and reasonable” payment amounts to the utility. In the utility regulation context, the phrase “just and reasonable” reflects the aim of “navigating the straits” between overcharging a utility’s customers and underpaying the utility for the public service it provides: Verizon Communications Inc. v. Federal Communications Commission , 535 U.S. 467 (2002), at p. 481; see also Northwestern Utilities Ltd. v. City of Edmonton , 1929 39 (SCC) , [1929] S.C.R. 186, at pp. 192-93.

[ 136 ] The methodology adopted by the Board to determine “just and reasonable” payments to Ontario Power Generation draws in part on the regulatory concept of “prudence”. Prudence is “a legal basis for adjudging the meeting of utilities’ public interest obligations, specifically in regard to rate proceedings”: Robert E. Burns et al., The Prudent Investment Test in the 1980s , report NRRI-84-16, The National Regulatory Research Institute, April 1985, at p. 20. The concept emerged in the early 20th century as a judicial response to the “mind-numbing complexity” of other approaches being used by regulators to determine “just and reasonable” amounts, and introduced a legal presumption that a regulated utility has acted reasonably: Verizon Communications , at p. 482. As Justice Brandeis famously explained in 1923:

The term prudent investment is not used in a critical sense. There should not be excluded from the finding of the base, investments which, under ordinary circumstances, would be deemed reasonable. The term is applied for the purpose of excluding what might be found to be dishonest or obviously wasteful or imprudent expenditures. Every investment may be assumed to have been made in the exercise of reasonable judgment, unless the contrary is shown. [Emphasis added.]

( State of Missouri ex rel. Southwestern Bell Telephone Co. v. Public Service Commission of Missouri , 262 U.S. 276 (1923), at p. 289, fn. 1, per Brandeis J., dissenting)

[ 137 ] The presumption of prudence is the starting point for the type of examination the Board calls a “prudence review”. In undertaking a prudence review, the Board applies a “well-established set of principles”:

• Decisions made by the utility’s management should generally be presumed to be prudent unless challenged on reasonable grounds.

• To be prudent, a decision must have been reasonable under the circumstances that were known or ought to have been known to the utility at the time the decision was made.

• Hindsight should not be used in determining prudence, although consideration of the outcome of the decision may legitimately be used to overcome the presumption of prudence.

• Prudence must be determined in a retrospective factual inquiry, in that the evidence must be concerned with the time the decision was made and must be based on facts about the elements that could or did enter into the decision at the time.

( Enersource Hydro Mississauga Inc. (Re) , 2012 LNONOEB 373 (QL), at para. 55, citing Enbridge Gas Distribution Inc. (Re) , 2002 LNONOEB 4 (QL), at para. 3.12.2.)

[ 138 ] This form of prudence review, including a presumption of prudence and a ban on hindsight, was endorsed by the Board and by the Ontario Court of Appeal as an appropriate method to determine “just and reasonable” rates in Enbridge Gas Distribution Inc. (Re) , at paras. 3.12.1 to 3.12.5, aff’d Enbridge Gas Distribution Inc. v. Ontario Energy Board (2006), 2006 10734 (ON CA) , 210 O.A.C. 4, at paras. 8 and 10-12 .

[ 139 ] In the case before us, however, the Board decided not to submit all costs to a prudence review. Instead, it stated that it would use two kinds of review. The first would apply to “forecast costs”, that is, those over which a utility retains discretion and can still be reduced or avoided. It explained in its reasons that it would review such costs using a wide range of evidence, and that the onus was on the utility to demonstrate that its forecast costs were reasonable:

When considering forecast costs, the onus is on the company to make its case and to support its claim that the forecast expenditures are reasonable. The company provides a wide spectrum of such evidence, including business cases, trend analysis, benchmarking data, etc. The test is not dishonesty, negligence, or wasteful loss; the test is reasonableness. And in assessing reasonableness, the Board is not constrained to consider only factors pertaining to [Ontario Power Generation]. The Board has the discretion to find forecast costs unreasonable based on the evidence — and that evidence may be related to the cost/benefit analysis, the impact on ratepayers, comparisons with other entities, or other considerations.

The benefit of a forward test period is that the company has the benefit of the Board’s decision in advance regarding the recovery of forecast costs. To the extent costs are disallowed, for example, a forward test period provides the company with the opportunity to adjust its plans accordingly. In other words, there is not necessarily any cost borne by shareholders (unless the company decides to continue to spend at the higher level in any event). [paras. 74-75]

[ 140 ] A different approach, the Board said, would be applied to those costs the company could not “take action to reduce”. These costs, sometimes called “committed costs”, represent binding commitments that leave a utility with no discretion about whether to make the payment. The Board explained that it evaluates these costs using a “prudence review”, which includes a presumption that the costs were prudently incurred:

Somewhat different considerations will come into play when undertaking an after-the-fact prudence review. In the case of an after-the-fact prudence review, if the Board disallows a cost, it is necessarily borne by the shareholder. There is no opportunity for the company to take action to reduce the cost at that point. For this reason, the Board concludes there is a difference between the two types of examination, with the after-the-fact review being a prudence review conducted in the manner which includes a presumption of prudence. [para. 75]

[ 141 ] In Enersource Hydro Mississauga Inc. (Re) , for example, the Board concluded that it had to conduct a prudence review when evaluating the costs that Enersource had already incurred:

This issue concerns expenditures which have largely already been incurred by the company. . . . Given that the issue concerns past expenditures which are now in dispute, the Board must conduct a prudence review. [para. 55]

[ 142 ] As the Board said in its reasons, the prudence review makes sense for committed costs because disallowing costs Ontario Power Generation cannot avoid, forces the utility to pay out of pocket for expenses it has already incurred. This could negatively affect Ontario Power Generation’s ability to operate, leading the utility to restructure its relationships with the financial community and its service providers, or even lead to bankruptcy: see Burns et al., at pp. 129-65. These outcomes would “increase capital costs and utility rates above the levels that would exist with a limited prudence penalty”, forcing Ontario consumers to pay higher electricity bills: Burns et al., at p. vi.

[ 143 ] The issue in this appeal therefore centres on the Board assessing all compensation costs in Ontario Power Generation’s collective agreements as adjustable forecast costs, without determining whether any of them were costs for which “[t]here is no opportunity for the company to take action to reduce” (para. 75). The Board did not actually call them forecast costs, but by saying that “collective agreements may make it difficult to eliminate positions quickly” and that “changes to union contracts . . . will take time” (paras. 346 and 352), the Board was clearly treating them as reducible in theory. Moreover, the fact that it failed to apply the prudence review it said it would apply to non-reducible costs confirms that it saw the collectively bargained commitments as adjustable.

[ 144 ] The Board did not explain why it considered compensation costs in collective agreements to be adjustable forecast costs, but the effect of its approach was to deprive Ontario Power Generation of the benefit of the Board’s assessment methodology that treats committed costs differently. In my respectful view, the Board’s failure to separately assess the compensation costs committed as a result of the collective agreements from other compensation costs, ignored not only its own methodological template, but labour law as well.

[ 145 ] Ontario Power Generation was a party to binding collective agreements with the Power Workers’ Union and the Society of Energy Professionals covering most of the relevant period. At the time of the application, it had already entered into a collective agreement with the Power Workers’ Union for the period of April 1, 2009 to March 31, 2012.

[ 146 ] Its collective agreement with the Society of Energy Professionals, which required resolution by binding mediation-arbitration in the event of contract negotiations disputes, expired on December 31, 2010. As a result of a bargaining impasse, the terms of a new collective agreement for January 1, 2011 to December 31, 2012 were imposed by legally binding arbitration: Ontario Power Generation v. Society of Energy Professionals , [2011] O.L.A.A. No. 117 (QL).

[ 147 ] The collective agreements with the Power Workers’ Union and the Society of Energy Professionals prescribed the compensation rates for staff positions held by represented employees, strictly regulated staff levels at Ontario Power Generation’s facilities, and limited the utility’s ability to unilaterally reduce its compensation rates and staffing levels. The collective agreement with the Power Workers’ Union, for example, stipulated that there would be no involuntary layoffs during the term of the agreement. Instead, Ontario Power Generation would be required either to relocate surplus staff or offer severance in accordance with rates set out in predetermined agreements between the utility and the union: “Collective Agreement between Ontario Power Generation Inc. and Power Workers’ Union”, April 1, 2009 to March 31, 2012, at art. 11.

[ 148 ] Similarly, Ontario Power Generation’s collective agreement with the Society of Energy Professionals severely limited the utility’s bargaining power and control over compensation levels. When the contract between Ontario Power Generation and the Society of Energy Professionals expired on December 31, 2010, the utility’s bargaining position had been that its sole shareholder, the Province of Ontario, had directed that there be a zero net compensation increase over the next two-year term. The parties could not reach an agreement and the dispute was therefore referred to binding arbitration as required by previous negotiations. The resulting award by Kevin M. Burkett provided mandatory across-the-board wage increases of 3 per cent on January 1, 2011, 2 per cent on January 1, 2012, and a further 1 per cent on April 1, 2012: Ontario Power Generation v. Society of Energy Professionals , at paras. 1, 9, and 28.

[ 149 ] The obligations contained in these collective agreements were immutable and legally binding commitments: Labour Relations Act, 1995 , s. 56. As a result, Ontario Power Generation was prohibited from unilaterally reducing the staffing levels, wages, or benefits of its unionized workforce. These agreements therefore did not just leave the utility “with limited flexibility regarding overall compensation rates or staffing levels”, as the majority notes (at para. 84), they made it illegal for the utility to alter the compensation and staffing levels of 90 per cent of its regulated workforce in a manner that was inconsistent with its commitments under the agreements.

[ 150 ] Instead, the Board, applying the methodology it said it would use for the utility’s forecast costs, put the onus on Ontario Power Generation to prove the reasonableness of its costs and concluded that it had failed to provide “compelling evidence” or “documentation or analysis” to justify compensation levels: para. 347. Had the Board used the approach it said it would use for costs the company had “no opportunity . . . to reduce”, it would have used an after-the-fact prudence review, with a rebuttable presumption that the utility’s expenditures were reasonable.

[ 151 ] Applying a prudence review to these compensation costs would hardly, as the majority suggests, “have conflicted with the burden of proof in the Ontario Energy Board Act, 1998 ”. To interpret the burden of proof in s. 78.1(6) of the Ontario Energy Board Act, 1998 so strictly would essentially prevent the Board from ever conducting a prudence review, notwithstanding that it has comfortably done so in the past and stated, even in its reasons in this case, that it would review committed costs using an “after-the-fact prudence review” which “includes a presumption of prudence”. Under the majority’s logic, however, since a prudence review always involves a presumption of prudence, the Board would not only be limiting its methodological flexibility, it would be in breach of the Act.

[ 152 ] The application of a prudence review does not shield the utility’s compensation costs from scrutiny. As the Court of Appeal observed, a prudence review

does not mean that the [Board] is powerless to review the compensation rates for [Ontario Power Generation’s] unionized staff positions or the number of those positions. In a prudence review, the evidence may show that the presumption of prudently incurred costs should be set aside, and that the committed compensation rates and staffing levels were not reasonable; however, the [Board] cannot resort to hindsight, and must consider what was known or ought to have been known at the time. A prudence review allows for such an outcome, and permits the [Board] both to fulfill its statutory mandate and to serve as a market proxy, while maintaining a fair balance between [Ontario Power Generation] and its customers. [para. 38]

[ 153 ] The majority’s suggestion (at para. 114) that “if the legislature had intended for costs under collective agreements to also be inevitably imposed on consumers, it would not have seen fit to grant the Board oversight of utility compensation costs”, is puzzling. The legislature did not intend for any costs to be “inevitably” imposed on consumers. What it intended was to give the Board authority to determine just and reasonable payment amounts based on Ontario Power Generation’s existing and proposed commitments. Neither collective agreements nor any other contractual obligations were intended to be “inevitably” imposed. They were intended to be inevitably considered in the balance. But it is precisely because of the unique nature of binding commitments that the Board said it would impose a different kind of review on these costs.

[ 154 ] It may well be that Ontario Power Generation has the ability to manage some staffing levels through attrition or other mechanisms that did not breach the utility’s commitments under its collective agreements, and that these costs may therefore properly be characterized as forecast costs. But no factual findings were made by the Board about the extent of any such flexibility. There is in fact no evidence in the record, nor any evidence cited in the Board’s decision, setting out what proportion of Ontario Power Generation’s compensation costs were fixed and what proportion remained subject to the utility’s discretion. The Board made virtually no findings of fact regarding the extent to which the utility could reduce its collectively bargained compensation costs. On the contrary, the Board, as Aitken J. noted, “lumped” all compensation costs together, acknowledged that reducing those in the collective agreements would “take time” and “be difficult”, and dealt with them as globally adjustable.

[ 155 ] Given that collective agreements are legally binding, it was unreasonable for the Board to assume that Ontario Power Generation could reduce the costs fixed by these contracts in the absence of any evidence to that effect. To use the majority’s words, these costs are “legal obligations that leave [the utility] with no discretion as to whether to make the payment in the future” (para. 82). According to the Board’s own methodology, costs for which “[t]here is no opportunity for the company to take action to reduce” are entitled to “a presumption of prudence”: para. 75.

[ 156 ] Disallowing costs that Ontario Power Generation is legally required to pay as a result of its collective agreements, would force the utility and the Province of Ontario, the sole shareholder, to make up the difference elsewhere. This includes the possibility that Ontario Power Generation would be forced to reduce investment in the development of capacity and facilities. And because Ontario Power Generation is Ontario’s largest electricity generator, it may not only threaten the “financial viability” of the province’s electricity industry, it could also imperil the assurance of reliable electricity service.

[ 157 ] The majority nonetheless assumes that the ongoing relationship between Ontario Power Generation and the unions should give the Board greater latitude in disallowing the collectively bargained compensation costs than it would have had if it applied a no-hindsight, presumption-of-prudence analysis. It also accepts the Board’s conclusion that Ontario Power Generation’s collectively bargained compensation costs may be “excessive”, and therefore concludes that the Board was reasonable in choosing to avoid the “prudence” test in order to so find. This approach finds no support even in the methodology the Board set out for itself for evaluating just and reasonable payment amounts.

[ 158 ] In my respectful view, selecting a test which is more likely to confirm an assumption that collectively bargained costs are excessive, misconceives the point of the exercise, namely, to determine whether those costs were in fact excessive. Blaming collective bargaining for what are assumed to be excessive costs, imposes, with respect, the appearance of an ideologically driven conclusion on what is intended to be a principled methodology based on a distinction between committed and forecast costs, not between costs which are collectively bargained and those which are not.

[ 159 ] I recognize that the Board has wide discretion to fix payment amounts that are “just and reasonable” and, subject to certain limitations, to “establish the . . . methodology” used to determine such amounts: O. Reg. 53/05, s. 6 , Ontario Energy Board Act, 1998 , s. 78.1. That said, once the Board establishes a methodology to determine what is just and reasonable, it is, at the very least, required to faithfully apply that approach: see TransCanada Pipelines Ltd. v. National Energy Board (2004), 2004 FCA 149 , 319 N.R. 171 (F.C.A.), at paras. 30-32 , per Rothstein J.A. This does not mean that collective agreements “supersede” or “trump” the Board’s authority to fix payment amounts; it means that once the Board selects a methodology for itself for the exercise of its discretion, it is required to follow it. Absent methodological clarity and predictability, Ontario Power Generation would be left in the dark about how to determine what expenditures and investments to make and how to present them to the Board for review. Wandering sporadically from approach to approach, or failing to apply the methodology it declares itself to be following, creates uncertainty and leads, inevitably, to needlessly wasting public time and resources in constantly having to anticipate and respond to moving regulatory targets.

[ 160 ] In disallowing $145 million of the compensation costs sought by Ontario Power Generation on the grounds that the utility could reduce salary and staffing levels, the Board ignored the legally binding nature of the collective agreements and failed to distinguish between committed compensation costs and those that were reducible. Whether or not one can fault the Board for failing to use a particular methodology, what the Board can unquestionably be analytically faulted for, is evaluating all compensation costs fixed by collective agreements as being amenable to adjustment. Treating these compensation costs as reducible was, in my respectful view, unreasonable.

[ 161 ] I would accordingly dismiss the appeal, set aside the Board’s decision, and, like the Court of Appeal, remit the matter to the Board for reconsideration in accordance with these reasons.

Appeal allowed, Abella J. dissenting.

Solicitors for the appellant: Stikeman Elliott, Toronto.

Solicitors for the respondent Ontario Power Generation Inc . : Torys, Toronto; Ontario Power Generation Inc ., Toronto .

Solicitors for the respondent the Power Workers’ Union, Canadian Union of Public Employees, Local 1000: Paliare Roland Rosenberg Rothstein, Toronto.

Solicitors for the respondent the Society of Energy Professionals: Cavalluzzo Shilton McIntyre Cornish, Toronto.

Solicitors for the intervener: Jay Shepherd Professional Corporation, Toronto.