Whiten v. Pilot Insurance Co., [2002] 1 S.C.R. 595, 2002 SCC 18

Daphne Whiten Appellant/Respondent on cross-appeal

v.

Pilot Insurance Company Respondent/Appellant on cross-appeal

and

The Insurance Council of Canada and

the Ontario Trial Lawyers Association Interveners

Indexed as: Whiten v. Pilot Insurance Co.

Neutral citation: 2002 SCC 18.

File No.: 27229.

2000: December 14; 2002: February 22.

Present: McLachlin C.J. and L’Heureux‑Dubé, Gonthier, Major, Binnie, Arbour and LeBel JJ.

on appeal from the court of appeal for ontario

Insurance – Insurer’s duty of good faith and fair dealing – Insurer contesting fire insurance claim in bad faith – Whether policy holder entitled to award of punitive damages ^__ Whether jury charge adequate – Whether jury award of $1 million in punitive damages should be restored.

Damages – Punitive damages – Insurer’s duty of good faith and fair dealing – Insurer contesting fire insurance claim in bad faith Whether policy holder entitled to award of punitive damages Whether jury award of $1 million in punitive damages should be restored.

The appellant and her husband discovered a fire in the addition to their house just after midnight in January 1994. They and their daughter fled the house wearing only their night clothes. It was minus 18 degrees Celsius. The husband gave his slippers to his daughter to go for help and suffered serious frostbite to his feet. The fire totally destroyed the home and its contents, including three cats. The appellant was able to rent a small winterized cottage nearby for $650 per month. The respondent insurer made a single $5,000 payment for living expenses and covered the rent for a couple of months or so, then cut off the rent without telling the family, and thereafter pursued a confrontational policy. The appellant’s family was in very poor financial shape. Ultimately this confrontation led to a protracted trial based on the respondent’s allegation that the family had torched its own home, even though the local fire chief, the respondent’s own expert investigator, and its initial expert all said there was no evidence whatsoever of arson. The respondent’s position was wholly discredited at trial and its appellate counsel conceded that there was no air of reality to the allegation of arson. The jury awarded compensatory damages and $1 million in punitive damages. A majority of the Court of Appeal allowed the appeal in part and reduced the punitive damages award to $100,000.

Held (LeBel J. dissenting on the appeal): The appeal should be allowed and the jury award of $1 million in punitive damages restored. The respondent’s cross-appeal against the award of any punitive damages should be dismissed.

Per McLachlin C.J. and L’Heureux‑Dubé, Gonthier, Major, Binnie and Arbour JJ.: The jury’s award of punitive damages, though high, was within rational limits. The respondent insurer’s conduct towards the appellant was exceptionally reprehensible. It forced her to put at risk her only remaining asset (the $345,000 insurance claim) plus $320,000 in costs that she did not have. The denial of the claim was designed to force her to make an unfair settlement for less than she was entitled to. The conduct was planned and deliberate and continued for over two years, while the financial situation of the appellant grew increasingly desperate. The jury evidently believed that the respondent knew from the outset that its arson defence was contrived and unsustainable. Insurance contracts are sold by the insurance industry and purchased by members of the public for peace of mind. The more devastating the loss, the more the insured may be at the financial mercy of the insurer, and the more difficult it may be to challenge a wrongful refusal to pay the claim.

The jury decided a powerful message of denunciation, retribution and deterrence had to be sent to the respondent and they sent it. The obligation of good faith dealing means that the appellant’s peace of mind should have been the respondent’s objective, and her vulnerability ought not to have been aggravated as a negotiating tactic. It is this relationship of reliance and vulnerability that was outrageously exploited by the respondent in this case.

An award of punitive damages in a contract case, though rare, is obtainable. It requires an “actionable wrong” in addition to the breach sued upon. Here, in addition to the contractual obligation to pay the claim, the respondent was under a distinct and separate obligation to deal with its policyholders in good faith. A breach of the contractual duty of good faith was thus independent of and in addition to the breach of contractual duty to pay the loss. The plaintiff specifically asked for punitive damages in her statement of claim and if the respondent was in any doubt about the facts giving rise to the claim, it ought to have applied for particulars.

The trial judge’s charge to the jury with respect to punitive damages should include words to convey an understanding of the following points: (1) Punitive damages are very much the exception rather than the rule, (2) imposed only if there has been high-handed, malicious, arbitrary or highly reprehensible misconduct that departs to a marked degree from ordinary standards of decent behaviour. (3) Where they are awarded, punitive damages should be assessed in an amount reasonably proportionate to such factors as the harm caused, the degree of the misconduct, the relative vulnerability of the plaintiff and any advantage or profit gained by the defendant, (4) having regard to any other fines or penalties suffered by the defendant for the misconduct in question. (5) Punitive damages are generally given only where the misconduct would otherwise be unpunished or where other penalties are or are likely to be inadequate to achieve the objectives of retribution, deterrence and denunciation. (6) Their purpose is not to compensate the plaintiff, but (7) to give a defendant his or her just desert (retribution), to deter the defendant and others from similar misconduct in the future (deterrence), and to mark the community’s collective condemnation (denunciation) of what has happened. (8) Punitive damages are awarded only where compensatory damages, which to some extent are punitive, are insufficient to accomplish these objectives, and (9) they are given in an amount that is no greater than necessary to rationally accomplish their purpose. (10) The jury should be told that while normally the state would be the recipient of any fine or penalty for misconduct, the plaintiff will keep punitive damages as a “windfall” in addition to compensatory damages. (11) Judges and juries in our system have usually found that moderate awards of punitive damages, which inevitably carry a stigma in the broader community, are generally sufficient. While the jury charge in this case was skeletal, it was upheld by the Court of Appeal (unanimous on this point) and, with hesitation, this Court should not allow the appeal on that ground.

As to quantum, the award of $1 million in punitive damages was more than this Court would have awarded, but was still within the high end of the range where juries are free to make their assessment.

Per LeBel J. (dissenting on the appeal): While the respondent’s bad faith in its handling of the claim, up to and during the trial, amply justifies awarding punitive damages, an award of $1 million – three times the compensation for loss of property – goes well beyond a rational and appropriate use of this kind of remedy. This case started as litigation based on a home insurance contract. The appellant suffered a loss and encountered obduracy and bad faith on the part of the respondent. There was no evidence that such conduct was a regular incident of the respondent’s way of running its business. Nor is such behaviour widespread in the Canadian insurance industry. The need for general deterrence here is far from clear. Concerns about industry practices should mainly be addressed through the appropriate regulatory and penal regimes, rather than through haphazard punitive damages awards.

The award fails the rationality test because its sole purpose is to punish bad faith and unfair dealing by the respondent. The award also fails the proportionality test because the punishment far exceeds whatever property or economic losses may have been caused by the non-performance of the contract. The Court of Appeal properly set the amount of punitive damages at a sum that is consistent with the nature and purpose of punitive damages in the law of torts. The amount appears reasonable and proportionate. It imposed significant punishment for the bad faith of the respondent without upsetting the proper balance between the compensatory and punitive functions of tort law. Predictability and consistency should be factored into situations where the nature of the damages suffered makes it difficult for a jury to determine a proper quantum. Setting punitive damages at amounts that do not significantly exceed the real economic loss would leave them in their proper place within the scheme of the law of torts. Where trials are held with a judge and jury, instructions on the range of awards would be useful. The jury should be instructed clearly that an award of general damages may also amount to all the punishment that is necessary in a given case.

Cases Cited

By Binnie J.

Applied: Hill v. Church of Scientology of Toronto, 1995 59 (SCC), [1995] 2 S.C.R. 1130; referred to: Vorvis v. Insurance Corp. of British Columbia, 1989 93 (SCC), [1989] 1 S.C.R. 1085; Wallace v. United Grain Growers Ltd., 1997 332 (SCC), [1997] 3 S.C.R. 701; Cassell & Co. v. Broome, [1972] A.C. 1027, aff’g [1971] 2 Q.B. 354; BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996); BMW of North America, Inc. v. Gore, 701 So.2d 507 (1997); Wilkes v. Wood (1763), Lofft. 1, 98 E.R. 489; Huckle v. Money (1763), 2 Wils. K.B. 206, 95 E.R. 768; Collette v. Lasnier (1886), 1886 54 (SCC), 13 S.C.R. 563; Rookes v. Barnard, [1964] A.C. 1129; Kuddus v. Chief Constable of Leicestershire Constabulary, [2001] 3 All E.R. 193; Uren v. John Fairfax & Sons Pty. Ltd. (1966), 117 C.L.R. 118; Taylor v. Beere, [1982] 1 N.Z.L.R. 81; Conway v. Irish National Teachers’ Organisation (1991), 11 I.L.R.M. 497; John v. MGN Ltd., [1997] Q.B. 586; Thompson v. Commissioner of Police of the Metropolis, [1997] 2 All E.R. 762; Lamb v. Cotogno (1987), 164 C.L.R. 1; XL Petroleum (N.S.W.) Pty. Ltd. v. Caltex Oil (Australia) Pty. Ltd. (1985), 155 C.L.R. 448; Australian Consolidated Press Ltd. v. Uren (1966), 117 C.L.R. 185; Whitfeld v. De Lauret & Co. (1920), 29 C.L.R. 71; Gray v. Motor Accident Commission (1998), 196 C.L.R. 1; M‘Comb v. Low (1873), 1 N.Z. Jur. 49; Donselaar v. Donselaar, [1982] 1 N.Z.L.R. 97; Daniels v. Thompson, [1998] 3 N.Z.L.R. 22; Cook v. Evatt (No. 2), [1992] 1 N.Z.L.R. 676; McLaren Transport Ltd. v. Somerville, [1996] 3 N.Z.L.R. 424; Aquaculture Corp. v. New Zealand Green Mussel Co., [1990] 3 N.Z.L.R. 299; Coloca v. B.P. Australia Ltd., [1992] 2 V.R. 441; L. v. Robinson, [2000] 3 N.Z.L.R. 499; Ellison v. L., [1998] 1 N.Z.L.R. 416; Auckland City Council v. Blundell, [1986] 1 N.Z.L.R. 732; Green v. Matheson, [1989] 3 N.Z.L.R. 564; McKenzie v. Attorney-General, [1992] 2 N.Z.L.R. 14; Dunlea v. Attorney-General, [2000] 3 N.Z.L.R. 136; W. v. W., [1999] 2 N.Z.L.R. 1; Cooper v. O’Connell, No. 85/90-96, 1997 Ireland S.C. Lexis; Day v. Woodworth, 54 U.S. (13 How.) 363 (1851); Fay v. Parker, 53 N.H. 342 (1872); Liebeck v. McDonald’s Restaurants, P.T.S., Inc., 1995 WL 360309; Pacific Mutual Life Insurance Co. v. Haslip, 499 U.S. 1 (1991); Honda Motor Co. v. Oberg, 512 U.S. 415 (1994); TXO Production Corp. v. Alliance Resources Corp., 509 U.S. 443 (1993); Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 121 S.Ct. 1678 (2001); M. (K.) v. M. (H.), 1992 31 (SCC), [1992] 3 S.C.R. 6; Denison v. Fawcett, 1958 119 (ON CA), [1958] O.R. 312; Robitaille v. Vancouver Hockey Club Ltd. (1981), 1981 532 (BC CA), 124 D.L.R. (3d) 228; Buxbaum (Litigation guardian of) v. Buxbaum, [1997] O.J. No. 5166 (QL); Glendale v. Drozdzik (1993), 1993 6867 (BC CA), 77 B.C.L.R. (2d) 106; Pollard v. Gibson (1986), 1 Y.R. 167; Joanisse v. Y. (D.) (1995), 1995 2783 (BC SC), 15 B.C.L.R. (3d) 224; Canada v. Lukasik (1985), 1985 1146 (AB QB), 18 D.L.R. (4th) 245; Wittig v. Wittig (1986), 1986 3081 (SK QB), 53 Sask. R. 138; Royal Bank of Canada v. W. Got & Associates Electric Ltd., 1999 714 (SCC), [1999] 3 S.C.R. 408; Central Trust Co. v. Rafuse, 1986 29 (SCC), [1986] 2 S.C.R. 147; Andrusiw v. Aetna Life Insurance Co. of Canada (2001), 2001 61004 (AB QB), 289 A.R. 1; Edwards v. Harris-Intertype (Canada) Ltd. (1983), 1983 1611 (ON SC), 40 O.R. (2d) 558, aff’d (1984), 1984 2070 (ON CA), 9 D.L.R. (4th) 319; Grenn v. Brampton Poultry Co. (1959), 1959 394 (ON CA), 18 D.L.R. (2d) 9; Starkman v. Delhi Court Ltd. (1960), 1960 413 (ON SC), 24 D.L.R. (2d) 152, aff’d (1961), 1961 195 (ON CA), 28 D.L.R. (2d) 269; Gastebled v. Stuyck (1973), 1973 2285 (FC), 12 C.P.R. (2d) 102, aff’d (1974), 1974 2447 (FCA), 15 C.P.R. (2d) 137; Paragon Properties Ltd. v. Magna Envestments Ltd. (1972), 1972 ALTASCAD 8, 24 D.L.R. (3d) 156; Rieger v. Burgess, 1988 209 (SK CA), [1988] 4 W.W.R. 577; Lauscher v. Berryere (1999), 1999 12242 (SK CA), 172 D.L.R. (4th) 439; Walker v. CFTO Ltd. (1987), 1987 126 (ON CA), 59 O.R. (2d) 104; Patenaude v. Roy (1994), 1994 6107 (QC CA), 123 D.L.R. (4th) 78; Recovery Production Equipment Ltd. v. McKinney Machine Co. (1998), 1998 ABCA 239, 223 A.R. 24; Mustaji v. Tjin (1996), 1996 1907 (BC CA), 30 C.C.L.T. (2d) 53; Québec (Curateur public) v. Syndicat national des employés de l'Hôpital St-Ferdinand (1994), 1994 6112 (QC CA), 66 Q.A.C. 1; Matusiak v. British Columbia and Yukon Territory Building and Construction Trades Council, [1999] B.C.J. No. 2416 (QL); Gerula v. Flores (1995), 1995 1096 (ON CA), 126 D.L.R. (4th) 506; Walker v. D’Arcy Moving & Storage Ltd. (1999), 1999 2808 (ON CA), 117 O.A.C. 367; United Services Funds (Trustees) v. Hennessey, [1994] O.J. No. 1391 (QL); Williams v. Motorola Ltd. (1998), 1998 5023 (ON CA), 38 C.C.E.L. (2d) 76; Procor Ltd. v. U.S.W.A. (1990), 1990 6637 (ON SC), 71 O.R. (2d) 410; Claiborne Industries Ltd. v. National Bank of Canada (1989), 1989 183 (ON CA), 69 O.R. (2d) 65; Horseshoe Bay Retirement Society v. S.I.F. Development Corp. (1990), 1990 8047 (BC SC), 66 D.L.R. (4th) 42; Kates v. Hall (1991), 1991 1127 (BC CA), 53 B.C.L.R. (2d) 322; Muir v. Alberta, 1996 7287 (AB QB), [1996] 4 W.W.R. 177; R. (L.) v. Nyp (1995), 25 C.C.L.T. (2d) 309; Weinstein v. Bucar, 1990 11072 (MB QB), [1990] 6 W.W.R. 615; Norberg v. Wynrib, 1992 65 (SCC), [1992] 2 S.C.R. 226; Nantel v. Parisien (1981), 18 C.C.L.T. 79; Lubrizol Corp. v. Imperial Oil Ltd. (1994), 1994 19680 (FC), 84 F.T.R. 197, rev’d (1996), 1996 4042 (FCA), [1996] 3 F.C. 40; Westbank Band of Indians v. Tomat, [1989] B.C.J. No. 1638 (QL).

By LeBel J. (dissenting on the appeal)

Ratych v. Bloomer, 1990 97 (SCC), [1990] 1 S.C.R. 940; Jacobi v. Griffiths, 1999 693 (SCC), [1999] 2 S.C.R. 570; Vorvis v. Insurance Corp. of British Columbia, 1989 93 (SCC), [1989] 1 S.C.R. 1085; Wallace v. United Grain Growers Ltd., 1997 332 (SCC), [1997] 3 S.C.R. 701; Edwards v. Law Society of Upper Canada, [2001] 3 S.C.R. 562, 2001 SCC 80; Cooper v. Hobart, [2001] 3 S.C.R. 537, 2001 SCC 79; Kamloops (City of) v. Nielsen, 1984 21 (SCC), [1984] 2 S.C.R. 2; Palsgraf v. Long Island R. Co., 162 N.E. 99 (1928); Canadian National Railway Co. v. Norsk Pacific Steamship Co., 1992 105 (SCC), [1992] 1 S.C.R. 1021; Hill v. Church of Scientology of Toronto, 1995 59 (SCC), [1995] 2 S.C.R. 1130; Cassell & Co. v. Broome, [1972] A.C. 1027; Andrews v. Grand & Toy Alberta Ltd., 1978 1 (SCC), [1978] 2 S.C.R. 229; Thornton v. Board of School Trustees of School District No. 57 (Prince George), 1978 12 (SCC), [1978] 2 S.C.R. 267; Arnold v. Teno, 1978 2 (SCC), [1978] 2 S.C.R. 287; ter Neuzen v. Korn, 1995 72 (SCC), [1995] 3 S.C.R. 674; Caron v. Chodan Estate (1992), 58 O.A.C. 173; Gray v. Alanco Developments Ltd., 1967 221 (ON CA), [1967] 1 O.R. 597; Howes v. Crosby (1984), 1984 2172 (ON CA), 45 O.R. (2d) 449.

Statutes and Regulations Cited

Courts of Justice Act, R.S.O. 1990, c. C.43, s. 118.

Rules of Civil Procedure, R.R.O. 1990, Reg. 194, Rule 25.06(9).

Authors Cited

Chapman, Bruce, and Michael Trebilcock. “Punitive Damages: Divergence in Search of a Rationale” (1989), 40 Ala. L. Rev. 741.

Feldthusen, Bruce. “Punitive Damages: Hard Choices and High Stakes”, [1998] N.Z. L. Rev. 741.

Feldthusen, Bruce. “Recent Developments in the Canadian Law of Punitive Damages” (1990), 16 Can. Bus. L.J. 241.

Halsbury’s Laws of Australia, vol. 9. Sydney: Butterworths, 1995.

Halsbury’s Laws of England, vol. 12(1), 4th ed. (reissue). By Lord Mackay of Clashfern. London: Butterworths, 1998.

Hibbert, Christopher. The Roots of Evil: A Social History of Crime and Punishment. Boston: Little, Brown, 1963.

Ireland. Law Reform Commission. Consultation Paper on Aggravated, Exemplary and Restitutionary Damages. Dublin: The Commission, 1998.

Linden, Allen M. Canadian Tort Law, 6th ed. Toronto: Butterworths, 1997.

McGregor on Damages, 16th ed. By Harvey McGregor. London: Sweet & Maxwell, 1997.

Ontario. Law Reform Commission. Report on Exemplary Damages. Toronto: The Commission, 1991.

Schlueter, Linda L., and Kenneth R. Redden. Punitive Damages, vol. 1, 4th ed. New York: Lexis, 2000.

United Kingdom. House of Commons, 6th series, Written Answers to Questions, November 9, 1999.

United Kingdom. Law Commission for England and Wales, Report 247. Aggravated, Exemplary and Restitutionary Damages, December 16, 1997.

Vogel, Paul G. Cohen Melnitzer’s Civil Procedure in Practice, vol. 1. Toronto: Carswell, 1989 (loose-leaf updated 1990, release No. 1).

Watson, Garry D., and Craig Perkins. Holmested and Watson: Ontario Civil Procedure, vol. 1. Toronto: Carswell, 1984 (loose-leaf updated 2001, release No. 5).

Weinrib, Ernest J. The Idea of Private Law. Cambridge, Mass.: Harvard University Press, 1995.

Windeyer, William John Victor. Lectures on Legal History, 2nd ed. rev. Sydney: Law Book, 1957.

APPEAL and CROSS-APPEAL from a judgment of the Ontario Court of Appeal (1999), 1999 3051 (ON CA), 42 O.R. (3d) 641, 170 D.L.R. (4th) 280, 117 O.A.C. 201, 32 C.P.C. (4th) 3, [1999] I.L.R. ¶ I-3659, [1999] O.J. No. 237 (QL), allowing in part the respondent’s appeal from a judgment of the Ontario Court (General Division) (1996), 1996 8109 (ON SC), 132 D.L.R. (4th) 568, 47 C.P.C. (3d) 229, [1996] O.J. No. 227 (QL). Appeal allowed, LeBel J. dissenting. Cross-appeal dismissed.

Gary R. Will and Anil Varma, for the appellant/respondent on cross-appeal.

Earl A. Cherniak, Q.C., and Kirk F. Stevens, for the respondent/appellant on cross-appeal.

Neil Finkelstein, Melanie L. Aitken and Russell Cohen, for the intervener the Insurance Council of Canada.

Robert B. Munroe, Andrew J. Spurgeon and Thomas P. Connolly, for the intervener the Ontario Trial Lawyers Association.

The judgment of McLachlin C.J. and L’Heureux‑Dubé, Gonthier, Major, Binnie and Arbour JJ. was delivered by

1 Binnie J. – This case raises once again the spectre of uncontrolled and uncontrollable awards of punitive damages in civil actions. The jury was clearly outraged by the high-handed tactics employed by the respondent, Pilot Insurance Company, following its unjustified refusal to pay the appellant’s claim under a fire insurance policy (ultimately quantified at approximately $345,000). Pilot forced an eight-week trial on an allegation of arson that the jury obviously considered trumped up. It forced her to put at risk her only remaining asset (the insurance claim) plus approximately $320,000 in legal costs that she did not have. The denial of the claim was designed to force her to make an unfair settlement for less than she was entitled to. The conduct was planned and deliberate and continued for over two years, while the financial situation of the appellant grew increasingly desperate. Evidently concluding that the arson defence from the outset was unsustainable and made in bad faith, the jury added an award of punitive damages of $1 million, in effect providing the appellant with a “windfall” that added something less than treble damages to her actual out-of-pocket loss. The respondent argues that the award of punitive damages is itself outrageous.

2 The appellant, Daphne Whiten, bought her home in Haliburton County, Ontario, in 1985. Just after midnight on January 18, 1994, when she and her husband Keith were getting ready to go to bed, they discovered a fire in the addition to their house. They and their daughter, who had also been upstairs, fled the house wearing only their night clothes. It was minus 18 degrees Celsius. Mr. Whiten gave his slippers to his daughter to go for help and suffered serious frostbite to his feet for which he was hospitalized. He was thereafter confined to a wheelchair for a period of time. The fire totally destroyed the Whitens’ home and its contents, including their few valuable antiques and many items of sentimental value and their three cats.

3 The appellant was able to rent a small winterized cottage nearby for $650 per month. Pilot made a single $5000 payment for living expenses and covered the rent for a couple of months or so, then cut off the rent without telling the family, and thereafter pursued a hostile and confrontational policy which the jury must have concluded was calculated to force the appellant (whose family was in very poor financial shape) to settle her claim at substantially less than its fair value. The allegation that the family had torched its own home was contradicted by the local fire chief, the respondent’s own expert investigator, and its initial expert, all of whom said there was no evidence whatsoever of arson. The respondent’s position, based on wishful thinking, was wholly discredited at trial. Pilot’s appellate counsel conceded here and in the Ontario Court of Appeal that there was no air of reality to the allegation of arson.

4 A majority of the Ontario Court of Appeal allowed the appeal in part and reduced the punitive damage award to $100,000. In my view, on the exceptional facts of this case, there was no basis on which to interfere with the jury award. The award, though very high, was rational in the specific circumstances disclosed in the evidence and within the limits that a jury is allowed to operate. The appellant was faced with harsh and unreasoning opposition from an insurer whose policy she had purchased for peace of mind and protection in just such an emergency. The jury obviously concluded that people who sell peace of mind should not try to exploit a family in crisis. Pilot, as stated, required the appellant to spend $320,000 in legal costs to collect the $345,000 that was owed to her. The combined total of $665,000 at risk puts the punitive damage awards in perspective. An award of $1 million in punitive damages is certainly at the upper end of a sustainable award on these facts but not beyond it. I would allow the appeal and restore the jury award of $1 million in punitive damages.

I. Facts

5 The facts surrounding the fire itself have already been briefly mentioned. The origin of the fire was never discovered but everyone who investigated the fire in the six months after it occurred concluded that it was accidental. The first persons to investigate the fire were the fire chief and firefighters called to the scene. The fire chief thought, and he was eventually shown to be correct, that the fire was caused at a single point of origin by a malfunctioning kerosene heater in the porch of the addition. This was where the fire was first observed and also the area which had sustained the most fire damage. The firefighters saw no evidence of arson and therefore they did not request the Fire Marshal’s office to investigate.

6 Pilot retained an experienced independent insurance adjuster, Derek Francis, to investigate the loss. Francis inspected the site and interviewed the Whitens, who freely acknowledged that they had both been unemployed and had financial difficulties. Francis also interviewed the firefighters about the speed at which the fire spread, a key indicator of arson. Both the physical evidence and the Whitens’ conduct satisfied Francis that the fire was accidental and on February 3, 1994 he reported to Pilot that “there is no suspicion of arson on behalf of the insureds or any members of their family”.

7 Francis made further investigations during which he determined that although the Whitens’ mortgage payments were in arrears, refinancing was being arranged. It appears that Pilot’s Senior Claims Examiner, Mr. Chris Porter, was already moving towards the conclusion that the claim should be disputed based on his suspicions of the family’s financial problems. In a letter dated February 25, 1994, Francis wrote to Pilot:

As outlined in my 2nd report with the physical evidence we have and the fact that the insured was attempting to arrange financing through another source and pay off the existing mortgage, there is little or no base [sic] to deny this claim. I certainly agree with your train of thought and if we did not have the physical evidence and the information from the insured’s solicitor that he was arranging financing for the Whitens, then my recommendations would certainly be opposite to what they are today. Unfortunately we must deal with the facts on hand and proceed with the adjustment accordingly in my opinion. [Emphasis added.]

8 Pilot did not agree that “there is little or no base [sic] to deny this claim”, although at that stage it had no evidence to support a defence of arson. It refused to accept Francis’s recommendations and decided to deny the claim. It did not tell Francis why it would not pay the claim and Francis in turn did not advise the Whitens of what was happening.

9 Pilot requested the Insurance Crime Prevention Bureau, a body set up by the insurance industry, to review the analysis of Pilot’s investigator. By letter dated February 25, 1994, the Bureau reported that “we wouldn’t have a leg to stand on as far as declining the claim”. Pilot, having asked for the opinion, then apparently decided that the Bureau’s evaluator was not in fact qualified to render an opinion. No one from Pilot testified as to why the claims examiner, and subsequently Pilot’s Branch Manager, Mr. Steven Carter, rejected this advice as well.

10 In March 1994, Pilot’s Head Office, with nothing to go on except some vague suspicions, instructed Francis to tell the landlord of the cottage that the appellant was renting that it was no longer going to pay the rent. Francis communicated this to the landlord but never told the appellant. We do not know why there was no communication. It was the depth of winter. The respondent’s Head Office made this decision with full knowledge that the Whitens were in desperate financial circumstances.

11 Pilot then instructed Francis to pursue other inquiries about the fire. Francis conducted these inquiries but still ruled out arson. On April 28, 1994, he confirmed his opinion to Pilot. In his reporting letter, he noted that he had come upon the Whitens unannounced and unexpected at the scene of the fire and found them sorting through the debris “trying to salvage anything that might have been left as a result of the fire”. He observed the appellant cleaning a small porcelain figure “with her fingers in an obvious attempt to salvage this item”. He reported that he “felt this genuine concern to try and see what could be salvaged now that the weather has afforded this opportunity [is] out of character for someone who might be involved in a suspicious fire”.

12 After receiving Francis’s report of April 28, 1994, Pilot removed Francis from the case and hired another adjuster, James Couch, who lived in Owen Sound, a couple of hundred kilometres distant. At trial, no one on behalf of Pilot testified as to why Pilot stopped using Francis.

13 Pilot also retained an engineering expert, Hugh Carter. His initial report was made on January 28, 1994. In that report, he concluded that the fire was accidental. He gave two further reports in which he stated the same opinion. Carter then received a letter dated May 4, 1994 from the respondent’s trial counsel, Donald Crabbe, which adverted to the arson theory:

One wonders whether the Whitens, even if they did not set the fire, sat back and allowed it to achieve a level that was convenient to them.

We need to be on top of this matter and to do it quickly. The other side has retained a lawyer and they are making noises of bad faith. The matter has to be revisited in its entirety, stripped down to the bare facts and rebuilt.

14 Hugh Carter concluded that he may have been misunderstood. He requested a meeting but did not get one at the time. The jury must have concluded that he had not provided the opinion his client wanted to hear.

15 The statement of claim was issued on May 27, 1994. On June 7, 1994, after a further site investigation, Carter did meet with Donald Crabbe and after the meeting, he reclassified the fire as “suspicious, possibly incendiary”. Pilot now concedes that Crabbe likely influenced Carter to change his opinion.

16 In its factum before this Court, Pilot also conceded that in addition to the Senior Claims Examiner and the Branch Manager, the latter’s “superior, George Hamilton (assistant to the Vice-President in charge of claims), [was] copied with all of the material on the file. Mr. Hamilton reported to Clifford Jones, Executive Vice President and Secretary” (para. 17). The misconduct was therefore not restricted to middle level management but was made known to the directing minds of the respondent company.

17 The attitude of the respondent and its counsel is apparent from Crabbe’s reporting letter dated June 9, 1994 to Pilot’s Chris Porter and Steven Carter (Pilot’s Branch Manager), parts of which read as follows:

The bottom line is that we have moved considerably with the upcoming engineer’s report towards successfully denying this claim. We still need more evidence, but we moved significantly in the right direction on June 7th.

18 It appears that all three people directing the respondent’s behaviour were agreed that the “right direction” was to deny the claim despite the lack of any evidence that the fire had been deliberately set. Crabbe continued:

In terms of the [appellant’s] punitive damage claim arising from bad faith, this is a cloud with a silver lining. First of all, it gives Hugh Carter a platform from which to discuss the evolution of his opinion. . . . More importantly in this concept of “silver lining” is that the claim renders admissible evidence as to the previous fires in which the Whitens were involved, when otherwise there was a considerable risk that the “similar fact” evidence would not be admitted as a significant enough pattern had perhaps not been established.

19 The reference to the two “previous fires” was firstly to a fire that occurred in a cottage owned by the Whitens’ son-in-law but rented out to a Mrs. Titro and secondly to another fire in another house previously occupied by Mrs. Titro. There was no apparent connection to the appellant or her family. At the Court of Appeal, Pilot conceded that evidence about these two fires was irrelevant and inadmissible. The reporting letter of June 9th continues:

You [Pilot] raise concerns that the other side has hired competent counsel and, frankly, I would not have it any other way and indeed it would be foolish to make any assumptions otherwise.

The jury must have asked itself why an insurer dealing in good faith with a policy holder would express “concern” to its own lawyer that she had hired competent counsel. Crabbe continued:

What we know is that counsel for the Plaintiff is advancing a claim based only on what the Plaintiff has told him. One can imagine the assertions of innocence and the exclusion of any reference to material that might be incriminating. I imagine the motivation was explored in terms of the financially bad times the family had fallen upon, but beyond that, the fire and all of its circumstances and the previous fires will not have been fully disclosed, if at all, to opposing counsel. By the time all of this evidence is disclosed, and coupled with the risk of a jury’s impression of it, competent counsel will view advancing the matter to trial as risky and [he] should be recommending a significant compromise, particularly in view of the fact that a trial will take quite some time. [Emphasis added.]

20 It was never explained how the Whitens stood to profit from torching their own home. The fair market value of their house was $157,000. The jury allowed $160,000. The mortgage still had to be paid out of the proceeds, leaving the appellant with only the existing equity in her home. Had the claim been paid promptly, the only financial effect on the appellant would have been to convert the roof over her family’s head into cash and oblige them to become renters. Selling the house would have had more or less the same financial impact. It defies common sense to think they would have risked so much — including their daughter’s safety, all of their possessions and their cats — for so little. Pilot now concedes (fairly, in my view) that the letter of June 9, 1994 “evinced an attitude which gave priority to Pilot’s interests at the expense of a dispassionate and fair approach to the interests of Mrs. Whiten”.

21 Thereafter, Pilot retained a forensic engineer, a fire investigator and a firefighter. Pilot did not disclose Francis’s exculpatory reports to any of these individuals, but instead, through Donald Crabbe, furnished them with information about the speed of the fire that the trial judge characterized as misleading if not inaccurate. The firefighter insisted that the fire was likely accidental but the other two experts gave opinions that provided some support for an arson defence. One of them, Richard Kooren, based his opinion on the existence of signs of a fire accelerant. Crabbe wrote on May 11, 1995:

Aside from the burn pattern under the washer, Richard Kooren sees liquid accelerant burn patterns on the annex floor which are not innocent. However, these observations are not made by [Pilot’s initial expert] Hugh Carter.

Pilot also conceded at the Court of Appeal that these inculpatory opinions were influenced by Crabbe.

22 The trial judge commented unfavourably about Crabbe’s role in this litigation. He felt that his “enthusiasm for his client’s case appears to have caused him to exceed the permissible limits which ought to confine a lawyer in the preparation of witnesses”. At the Court of Appeal and in this Court, Pilot conceded that these comments were justified, but added:

. . . Pilot, not its counsel, made the decision to deny the claim and Pilot was fully aware, because it was a recipient of the letters, of counsel’s “enthusiasm”. Pilot recognizes that it bears the responsibility for what occurred.

23 The appellant was reinstated in her nursing position in July 1994 and received about $40,000 in back pay in September 1994.

24 In the spring of 1995 the Whitens, in an attempt to satisfy Pilot that they did not set the fire, offered to take a polygraph test administered by an expert selected by Pilot. This was apparently accepted by the jury as a good faith offer made to allay Pilot’s suspicions. Pilot refused, without giving any reasons.

25 The Whitens lived in a small community. People were aware that their home was not being rebuilt because the insurer was alleging arson. The stigma persisted. Pilot continued to allege arson throughout the trial. Pilot now concedes that the evidence as a whole unequivocally demonstrates that the fire was accidental.

II. Judicial History

A. Ontario Court (General Division) (1996), [1996 8109 (ON SC)](https://www.canlii.org/en/on/onsc/doc/1996/1996canlii8109/1996canlii8109.html), 132 D.L.R. (4th) 568

26 This action was tried before Matlow J. and a jury. His instruction to the jury on the issue of punitive damages was skeletal but occasioned no objection from either counsel:

And finally, if you determine that Pilot’s defence of arson failed and that Pilot breached the provision of the policy of insurance by denying the plaintiff’s claim, you must then go on to determine whether the plaintiff is entitled, as well, to recover punitive damages. Punitive damages can be awarded in certain circumstances to serve as a punishment. In this case, depending on your finding of fact, punitive damages can be awarded to deter Pilot and other insurers from engaging in improper conduct in dealing with the claims of their insureds.

Punitive damages, unlike the other types of damages claimed in this case, are not intended to compensate the plaintiff for her loss. If they are awarded, they will constitute a windfall for the plaintiff and a penalty for Pilot.

Before you may properly make an award of punitive damages, Pilot’s defence of arson must fail and you must be satisfied that the plaintiff has proven that Pilot failed to deal with her claim in good faith and instead dealt with this in a malicious, high-handed, arbitrary or capricious manner, and that Pilot’s conduct warrants the imposition of a penalty.

27 After the charge, the jury returned with the following question:

Dear Justice Matlow, we are having difficulty in agreements pertaining to assessing the amount of the claim for punitive damages. Would you be able to provide us some guidelines to help us arrive at a consensus. Thank you, the jury.

28 Counsel were consulted and agreed with each other that no dollar amounts should be mentioned. Pilot’s counsel told the trial judge, “in terms of suggesting amounts or anything, I think it ought not to occur”. In these circumstances, the trial judge responded to the jury by repeating that punitive damages were in their discretion:

Members of the jury, I have considered the question that you sent to me and I don’t know that I can really be of all that much help to you. All that I can say to you is that punitive damages are in the discretion of the jury. You have to be fair and reasonable to both sides, and apart from that, there’s not much more or anything more that I can tell you. It is not surprising that it is difficult to arrive at a consensus. I urge you to keep talking to each other and endeavour to find what that magic figure should be.

29 The jury awarded $318,252.32 in compensatory damages and $1 million in punitive damages. In his endorsement granting judgment in accordance with the jury’s verdict, Matlow J. awarded pre-judgment interest and costs on a solicitor and own client scale, his intention being “to provide for the plaintiff’s fullest possible indemnification for her reasonable costs of this action” (p. 570). (The appellant’s costs were ultimately fixed at $317,658.92.)

30 Matlow J. then made a number of observations about the jury’s award of punitive damages. He said that although it was “very high and perhaps without precedent, [it] is not perverse but is entirely reasonable in light of all of the evidence” (p. 572). He noted that the defendant continued to deny the claim even after its own adjuster recommended that it be paid. “[T]he defendant relied on a few suspicious circumstances that were later clarified adequately by the plaintiff in order to press on with an ill-founded defence” (p. 572). “As a result, the plaintiff, who was already in poor financial condition, was required to endure the indignity of having to make temporary living arrangements without the benefit of the insurance coverage for which she had paid premiums to the defendant” (p. 572). She was also required to undertake litigation to secure the relief to which she was entitled, including a trial which took approximately two months to complete. “In light of the defendant’s admission that its net worth was approximately $231 million, I cannot take issue with the jury’s conclusion that a very substantial award for punitive damages was required to punish the defendant and to effectively send the implied reminder to the defendant and to other insurers that they owe their insureds a duty of good faith in responding to claims made under policies of insurance issued by them” (p. 572).

B. Ontario Court of Appeal (1999), [1999 3051 (ON CA)](https://www.canlii.org/en/on/onca/doc/1999/1999canlii3051/1999canlii3051.html), 42 O.R. (3d) 641

(1) Laskin J.A. (dissenting in part)

31 On the issue of punitive damages, Laskin J.A. noted that in Vorvis v. Insurance Corp. of British Columbia, 1989 93 (SCC), [1989] 1 S.C.R. 1085, and again in Wallace v. United Grain Growers Ltd., 1997 332 (SCC), [1997] 3 S.C.R. 701, this Court held that in a breach of contract case, punitive damages were only available if, apart from the breach sued upon, the defendant has committed an “independent actionable wrong”. Pilot argued that a breach of its duty of good faith was no more than another breach of the same contractual obligations. Laskin J.A. rejected this argument. He said the “obligation to act in good faith is separate from the insurer’s obligation to compensate its insured for a loss covered by the policy” (p. 650). In his view, breach of the former “is a separate or independent wrong from the wrong for which compensation is paid” (p. 650). He said that if the decision in Vorvis makes it necessary he “would be prepared to hold that an insure[r] has a duty in tort of good faith towards its insureds” (p. 652), a position taken by some Australian and American courts. In his view “[t]here was overwhelming evidence in this case from which the jury could reasonably conclude that Pilot’s handling of the Whitens’ claim was so malicious or vindictive or so reprehensible or high-handed that an award of punitive damages was warranted” (p. 653).

32 With respect to the $1 million quantum, Laskin J.A. rejected Pilot’s submissions that the jury likely inflated the award because of errors made by the trial judge in his charge, or that in any event, the award was much too high. While the award of $1 million was high, the jury’s decision was entitled to deference reinforced in this case by the trial judge’s endorsement of the award as “entirely reasonable”. In Laskin J.A.’s view, Pilot’s conduct was “exceptionally reprehensible”. The deterrence objective is particularly important in first party insurance cases given the vast number of claims handled by insurers every year. Moreover, to be meaningful, an award of punitive damages “must sting”; it must not be perceived as a mere licence fee or as a cost of doing business. Both courts and legislatures, in recent years, have increased the amount of fines for companies which act irresponsibly or contrary to the public interest. This trend reflects an acknowledgement that “larger fines are needed to deter and punish companies for socially unacceptable behaviour” (p. 661).

(2) Finlayson J.A. (Catzman J.A. concurring)

33 Finlayson J.A. agreed with Laskin J.A.’s reasons and conclusions on the first issue, namely, whether Ms. Whiten was entitled to an award of punitive damages. However, he did not agree with Laskin J.A. that the $1 million award was not excessive. Finlayson J.A. said that, while he was “not entirely happy with the trial judge’s charge to the jury” (p. 661), he did not propose to justify his intervention on any other basis than his belief that the award “is simply too high” (p. 661).

34 Finlayson J.A. reviewed the facts of and awards made in other cases to illustrate his concern about quantum. In all of these cases, the defendant insurer was found to have acted unacceptably; in some the plaintiff received solicitor‑client costs; in others he or she did not. None of the awards exceeded $50,000. The cases relied upon by the plaintiff went beyond bad faith and reflected the courts’ concern with “objectionable corporate policies” (p. 663). Moreover, in three of these cases the trial judge emphasized the need to force tortfeasors to disgorge profits flowing from their actions (pp. 661-65). In the case at bar, according to Finlayson J.A., there was no evidence that Pilot’s unacceptable conduct “was the product of a corporate strategy” (p. 665) nor did it profit from its actions. “Rather, it appears to have been an isolated instance for which the appellant’s trial counsel should take full responsibility, both for the manner in which the claim was processed and because of the way that the trial was conducted. This certainly was the view of the trial judge” (p. 665).

35 In the view of Finlayson J.A., an award of $100,000 would act as a sufficient deterrent to this and other insurers and would cause the former to take the steps necessary to ensure that “in future it is properly apprised of the nature and kind of the defence its claims adjusters and counsel are advancing” (p. 666).

III. Analysis

36 Punitive damages are awarded against a defendant in exceptional cases for “malicious, oppressive and high-handed” misconduct that “offends the court’s sense of decency”: Hill v. Church of Scientology of Toronto, 1995 59 (SCC), [1995] 2 S.C.R. 1130, at para. 196. The test thus limits the award to misconduct that represents a marked departure from ordinary standards of decent behaviour. Because their objective is to punish the defendant rather than compensate a plaintiff (whose just compensation will already have been assessed), punitive damages straddle the frontier between civil law (compensation) and criminal law (punishment).

37 Punishment is a legitimate objective not only of the criminal law but of the civil law as well. Punitive damages serve a need that is not met either by the pure civil law or the pure criminal law. In the present case, for example, no one other than the appellant could rationally be expected to invest legal costs of $320,000 in lengthy proceedings to establish that on this particular file the insurer had behaved abominably. Over-compensation of a plaintiff is given in exchange for this socially useful service.

38 Nevertheless, the hybrid nature of punitive damages offends some jurists who insist that legal remedies should belong to one jurisprudential field or the other. That is one major aspect of the controversy, often framed in the words of Lord Wilberforce’s comments, dissenting, in Cassell & Co. v. Broome, [1972] A.C. 1027 (H.L.), at p. 1114:

It cannot lightly be taken for granted, even as a matter of theory, that the purpose of the law of tort is compensation, still less that it ought to be, an issue of large social import, or that there is something inappropriate or illogical or anomalous (a question-begging word) in including a punitive element in civil damages, or, conversely, that the criminal law, rather than the civil law, is in these cases the better instrument for conveying social disapproval, or for redressing a wrong to the social fabric, or that damages in any case can be broken down into the two separate elements. As a matter of practice English law has not committed itself to any of these theories: it may have been wiser than it knew.

39 A second major aspect of the controversy surrounding punitive damages is related to the quantum. Substantial awards are occasionally assessed at figures seemingly plucked out of the air. The usual procedural protections for an individual faced with potential punishment in a criminal case are not available. Plaintiffs, it is said, recover punitive awards out of all proportion to just compensation. They are subjected, it is said, to “palm tree justice”: Cassell, supra, at p. 1087. They are handed a financial windfall serendipitously just because, coincidentally with their claim, the court desires to punish the defendant and deter others from similar outrageous conduct. Defendants on the other hand say they suffer out of all proportion to the actual wrongs they have committed. Because the punishment is tailored to fit not only the “crime” but the financial circumstances of the defendant (i.e., to ensure that it is big enough to “sting”), defendants complain that they are being punished for who they are rather than for what they have done. The critics of punitive awards refer in terrorem to the United States experience where, for example, an Alabama jury awarded $4 million in punitive damages against a BMW dealership for failure to disclose a minor paint job to fix a cosmetic blemish on a new vehicle in BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996). In 1994, a jury in New Mexico awarded 81-year-old Stella Liebeck $160,000 in compensatory damages and $2.7 million in punitive damages against McDonald’s Restaurants for burns resulting from a spilled cup of coffee, notwithstanding that she tried to open the cup while balancing it on her lap in the passenger seat of a car (Liebeck v. McDonald’s Restaurants, P.T.S., Inc., 1995 WL 360309 (D. N.M.)). Critics of punitive damages warn against an “Americanization” of our law that, if adopted, would bring the administration of justice in this country into disrepute.

40 These are serious concerns, but in fact, the punitive damage controversies have little if anything to do with Americanization of our law. Jury awards of punitive damages in civil actions have a long and important history in Anglo-Canadian jurisprudence. They defy modern attempts at neat classification of remedies. The jury is invited to treat a plaintiff as a public interest enforcer as well as a private interest claimant. Almost 240 years ago, government agents broke into the premises of a Whig member of Parliament and pamphleteer, John Wilkes, to seize copies of a publication entitled The North Briton, No. 45, which the Secretary of State regarded as libellous. Lord Chief Justice Pratt (later Lord Camden L.C.) on that occasion swept aside the government’s defence. “If such a [search] power is truly invested in a Secretary of State”, he held, “and he can delegate this power, it certainly may affect the person and property of every man in this kingdom, and is totally subversive of the liberty of the subject”. As to punitive damages, he affirmed that:

[A] jury have it in their power to give damages for more than the injury received. Damages are designed not only as a satisfaction to the injured person, but likewise as a punishment to the guilty, to deter from any such proceeding for the future, and as a proof of the detestation of the jury to the action itself.

(Wilkes v. Wood (1763), Lofft. 1, 98 E.R. 489 (K.B.), at pp. 498-99)

41 Long before the days of Lord Pratt C.J., the related idea of condemning a defendant to a multiple of what is required for compensation (in the present appeal, as stated, the punitive damages were roughly triple the award of compensatory damages) reached back to the Code of Hammurabi, Babylonian law, Hittite law (1400 B.C.), the Hindu Code of Manu (200 B.C.), ancient Greek codes, the Ptolemaic law in Egypt and the Hebrew Covenant Code of Mosaic law (see Exodus 22:1 “If a man shall steal an ox, or a sheep, and kill it, or sell it; he shall restore five oxen for an ox, and four sheep for a sheep”). Roman law also included provisions for multiple damages. Admittedly, in these early systems, criminal law and civil law were not always clearly differentiated. The United States Supreme Court in BMW, supra, referred at p. 581 to “65 different enactments [in English statutes] during the period between 1275 and 1753 [that] provided for double, treble, or quadruple damages”.

42 Even in terms of quantum, the use of punitive damages in the eighteenth century was aggressive. In Huckle v. Money (1763), 2 Wils. K.B. 206, 95 E.R. 768, the journeyman Huckle (who had actually printed the pamphlet The North Briton, No. 45 at issue in Wilkes, supra) won a cause of action for trespass, assault and false imprisonment and received 300 pounds in damages from the jury despite the comfortable and short six-hour duration of his confinement. The government’s motion for a new trial on the basis that the award was “outrageous” was denied, even though actual damages totalled only 20 pounds (i.e., a multiplier of 15) (p. 768). The Lord Chief Justice, in introducing the expression “exemplary damages”, thought there was no precedent for judges “intermeddling” with damages awarded by juries.

43 The three objectives identified by Lord Chief Justice Pratt, in Wilkes, supra – punishment, deterrence and denunciation (“proof of the detestation”) – are with us still, even though some scholarly critics have argued that these rationales “have very particular and divergent implications” that occasionally wind up undermining each other: B. Chapman and M. Trebilcock, “Punitive Damages: Divergence in Search of a Rationale” (1989), 40 Ala. L. Rev. 741, at p. 744. No doubt, as a matter of language, the word “punishment” includes both retribution and denunciation, and the three objectives should perhaps better be referred to as retribution, deterrence and denunciation.

44 The notion of private enforcers (or “private Attorneys General”), particularly where they act for personal gain, is worrisome unless strictly controlled. Thus, while the availability of punitive damages in Canada was affirmed early on by this Court in Collette v. Lasnier (1886), 1886 54 (SCC), 13 S.C.R. 563, a patent case, they were not widely awarded until the 1970s. Since then the awards have multiplied in number and escalated in amount. A report on punitive damages by the Ontario Law Reform Commission, issued in 1991, which examined research begun in 1989, predicted limited and principled development in the law of punitive damages in Canada: Ontario Law Reform Commission, Report on Exemplary Damages (1991), at pp. 93 and 98. By 1998, the report’s research director, Dean Bruce Feldthusen, conceded that the law was “certainly developing quite differently in Canada than one would have predicted only a short time ago” and that “many of the doctrinal pillars on which the Report’s predictions of limited and principled development in the law governing punitive damages were based have since cracked or collapsed”: B. Feldthusen, “Punitive Damages: Hard Choices and High Stakes”, [1998] N.Z. L. Rev. 741, at p. 742. Contrary to expectations, the awards were much larger, more frequent, appeared to rely more often on the defendant’s wealth in support, and included more high profile jury awards. The kinds of causes of action had expanded; punitive damages were the “norm” and had “proliferated” in actions in sexual battery, were now “clearly available” for breach of fiduciary duty, and “persisted” in contract actions. Prior criminal convictions, he concluded, no longer automatically barred punitive awards. He added, “[p]erhaps most significantly, the courts seem to have accepted general deterrence, not retributive punishment, as the dominant purpose behind punitive damage awards in a number of important decisions” (p. 742).

45 This Court more recently affirmed a punitive damage award of $800,000 in Hill, supra. On that occasion some guidelines were set out to keep this remedy within reasonable limits. The Court on this occasion has an opportunity to clarify further the rules governing whether an award of punitive damages ought to be made and if so, the assessment of a quantum that is fair to all parties.

46 It is convenient at this point to note how other common law jurisdictions have addressed the problem of disproportionate awards of punitive damages.

A. England

47 In Rookes v. Barnard, [1964] A.C. 1129, and Cassell, supra, the House of Lords sought to significantly limit the availability of punitive damages. In Lord Devlin’s leading speech in Rookes, he distinguished “exemplary” (punitive) and “aggravated” (compensatory) damages and classified many purportedly punitive awards in the law books as disguised compensation. Second, he reduced the availability of punitive damages to two common law categories of causes of action, namely (1) “oppressive, arbitrary, or unconstitutional action by the servants of the government” (p. 1226) and (2) where the defendant’s conduct is calculated to make a profit. To these he added a third permissible category where punitive damages have express statutory authorization. Since Rookes was decided, the categories test has persisted despite various academic and judicial assaults, but later cases have somewhat loosened the categories’ application (by, for example, including police wrongdoing in category one). More recently, in June of last year, the House of Lords opened up the categories themselves to further evolution: Kuddus v. Chief Constable of Leicestershire Constabulary, [2001] 3 All E.R. 193.

48 The attempt in Rookes to narrow the availability of punitive damages by a categorical approach, has been criticized in both academic and judicial forums. The Court of Appeal in England attempted to undermine the new doctrine in Broome v. Cassell & Co., [1971] 2 Q.B. 354; however, the attempt was crushed on further appeal of its decision to the House of Lords, [1972] A.C. 1027. The Law Commission for England and Wales concluded recently that the limitations comported neither with “sound principle” nor “sound policy” (Report 247, Aggravated, Exemplary and Restitutionary Damages (1997), at para 1.2). Canada, Australia, New Zealand, and Ireland have all explicitly rejected the “category” approach of Rookes: Vorvis, supra, at pp. 1104-6; Uren v. John Fairfax & Sons Pty. Ltd. (1966), 117 C.L.R. 118 (Austl. H.C.); Taylor v. Beere, [1982] 1 N.Z.L.R. 81 (C.A.); Conway v. Irish National Teachers’ Organisation, [1991] 11 I.L.R.M. 497 (S.C.). Category one has been criticized on the grounds that corporations and individuals can exercise enormous power in an outrageous way just as governmental actors do. Criticisms have been especially vigorous against category two which strikes many as illogical. Commentators wonder why conduct motivated by economic profit is punished while that inspired “only” by malice escapes. (Lord Nicholls lends support to both these views in Kuddus, supra, at paras. 66-67; see also Taylor J. in Uren, supra, at pp. 132-39; and Richardson J. in Taylor, supra, at p. 92.)

49 The study of punitive damages by the Law Commission for England and Wales was extensive, first seeking public consultation with two papers in 1993 and 1995, and finally reporting in 1997. It recommended that punitive damages should be retained and expanded, while also emphasizing that punitive damage awards should be “consistent, moderate and proportionate” (Part 5, at para. 82). The Law Commission was guided by five principles in formulating their specific recommendations for punitive damages (Part. 1, at para. 17):

First, exemplary damages should be an exceptional remedy, rarely-awarded and reserved for the most reprehensible examples of civil wrongdoing which would otherwise go unpunished by the law. Secondly, their availability (and assessment) must be placed on a clear, principled basis. Thirdly, although flexibility is necessary, unnecessary uncertainty as to the availability and assessment of the remedy must be avoided. Fourthly, defendants must not be unfairly prejudiced. Fifthly, the impact on the administration and funding of civil justice should not be adverse.

The specific proposals have not yet been accepted by the U.K. Parliament which is said to be awaiting further judicial development of the law in that country: House of Commons, Written Answers to Questions, November 9, 1999, col. 502.

50 At the present time, the English courts are developing a variety of principles and guidelines to control the award of punitive damages: Halsbury’s Laws of England (4th ed. 1998), vol. 12(1), at paras. 1115-17; McGregor on Damages (16th ed. 1997), at paras. 461-70; Law Commission for England and Wales, supra, Part IV. Some of these are substantive and some are procedural. In substantive terms, it is now well established that the plaintiff and not somebody else must be the victim of the defendant’s wrongful conduct; restraint should be exercised in the award of punitive damages; and punitive damages should be awarded under the Rookes approach “if, but only if” compensatory damages are inadequate to punish the defendant (which means that punitive damages are a “topping up” award and a remedy of last resort). Relevant factors include the reprehensibility of the defendant’s conduct; the existence of multiple plaintiffs or multiple defendants; the defendant’s good faith; and whether the defendant has already been punished under criminal law (this may be an absolute bar). The parties’ means are relevant to the assessment.

51 In terms of procedure, the English courts have made important changes. In John v. MGN Ltd., [1997] Q.B. 586, the Court of Appeal ruled that a jury’s attention can be directed to previous awards that the Court of Appeal has either approved or substituted for a jury award as well as to compensatory damage scales. The Court of Appeal also approved the practice of judges and counsel indicating an appropriate award level. The Court of Appeal felt that the process of mentioning figures, “far from developing into an auction”, would “induce a mood of realism” on both sides (p. 615). In Thompson v. Commissioner of Police of the Metropolis, [1997] 2 All E.R. 762, the Court of Appeal went further still and approved “guideline” brackets for both compensatory and punitive damages. The Court of Appeal set down general guiding principles, and minimum and ceiling amounts specific to the cause of action at issue, thus permitting brackets in the consideration of punitive damages claims. The Court of Appeal’s guiding principles for jury instruction included (1) the familiar “if, but only if” test; (2) pointing out that aggravated damages, because they compensate for injuries due to outrageous defendant behaviour, already include a measure of punishment, albeit incidental; (3) punitive damages are a “windfall” to the plaintiff; and (4) punitive damages against a public authority may mean that the sums are not available to be spent on behalf of the general public (i.e., punitive damages against the state come at the general public’s expense) (p. 776). In the specific case of an action of a malicious prosecution and false imprisonment against police, for example, the Court of Appeal selected brackets (to be adjusted for inflation) for the punitive damages that are “unlikely to be less than £5,000” as otherwise the conduct probably does not justify any punitive damages award. The “conduct must be particularly deserving of condemnation for an award of as much as £25,000 to be justified and the figure of £50,000 should be regarded as the absolute maximum, involving directly officers of at least the rank of superintendent” (p. 776). The Court of Appeal in Thompson also observed that a useful rule of thumb is that the total sum of damages (including punitive damages) should not exceed three times the basic compensatory damages (p. 777).

B. Australia

52 The High Court of Australia takes the view that the purposes of punitive damages are (1) deterrence of the wrongdoer, “other like-minded persons”, and generally of “conduct of the same reprehensible kind”; and (2) the social function of “teach[ing] a wrong-doer that tort does not pay” (Lamb v. Cotogno (1987), 164 C.L.R. 1, at p. 10; XL Petroleum (N.S.W.) Pty. Ltd. v. Caltex Oil (Australia) Pty. Ltd. (1985), 155 C.L.R. 448, at p. 471, per Brennan J.). Punitive damages “serve to mark the court’s condemnation” and “disapprobation” of the defendant’s behaviour (Lamb, supra, at pp. 10 and 13). Punitive damages require something more than a jury’s “mere disapproval” of the defendant’s conduct (Uren, supra, at p. 153). The High Court has endorsed moderation and proportion in punitive awards; proportionality is relative to the circumstances of the case, and there is “no necessary proportionality” between the assessment of compensatory and punitive damages (XL Petroleum (N.S.W.) Pty., supra, per Gibbs C.J., at p. 463, and Brennan J., at p. 471; Lamb, supra).

53 In 1966 in the companion cases of Uren, supra, and Australian Consolidated Press Ltd. v. Uren (1966), 117 C.L.R. 185, the High Court declined to adopt Rookes’s “categories” of restrictions on punitive damage awards. Punitive damages are available for “conscious wrongdoing in contumelious disregard of another’s rights” (Whitfeld v. De Lauret & Co. (1920), 29 C.L.R. 71 (Austl. H.C.), at p. 77, per Knox C.J.). Punitive damages are awarded “rarely” but in a wider range of torts than those prescribed in Rookes, supra. They are most commonly awarded for intentional torts but their availability is not precluded merely because the cause of action is framed in negligence if the wrongdoing is wanton, reckless or outrageous (Gray v. Motor Accident Commission (1998), 196 C.L.R. 1 (Austl. H.C.), at pp. 9 and 28; Lamb, supra). Thus punitive damages have been awarded for trespass to chattels, trespass to land, trespass to the person, deceit, reckless negligence, and defamation (Gray, supra, at pp. 27-28); and see generally, Halsbury’s Laws of Australia (1995), vol. 9, 1995, “Damages”, at paras. 135-505. The defendant’s misconduct, and not the cause of action, thus governs the availability of punitive damages.

54 With respect to the potential duplication of the punitive function of the criminal law, the High Court has ruled that the “infliction of substantial punishment for what is substantially the same conduct as the conduct which is the subject of the civil proceeding” bars the award of punitive damages (Gray, supra, at p. 14). The dissent on this point by Callinan J. is closer to the Canadian position at pp. 50-51:

The fact of the imposition of punishment and its extent and impact on the defendant will always be relevant factors, probably on most occasions the major and decisive factors. They may not however be conclusive ones for all cases. . . . A court would also be entitled to take into account that lesser punishments may have been, or might be imposed as a consequence of the acceptance of a lesser plea, the availability (for what might be sound policy reasons in and for the purposes of the criminal law) of a small penalty only, the desirability of the less condemnatory process by way of civil rather than criminal proceedings, the need to encourage compliance with the law, and the fact that the possibility of any criminal sanction is illusory.

C. New Zealand

55 Punitive damages in tort have always been part of New Zealand’s common law and have been recognized repeatedly by its Supreme Court and Court of Appeal. See, for example, M‘Comb v. Low (1873), 1 N.Z. Jur. 49 (punitive damages award in malicious prosecution). After Rookes, supra, New Zealand affirmed the judicial power to mark high-handed and heinous conduct in contumelious disregard of another’s rights through the award of punitive damages without limitation to Rookes’s narrow categories (Taylor, supra; Donselaar v. Donselaar, [1982] 1 N.Z.L.R. 97 (C.A.); and Daniels v. Thompson, [1998] 3 N.Z.L.R. 22 (C.A.)).

56 The New Zealand courts have permitted punitive damage awards in a wide range of causes of action in both common law and equity (Cook v. Evatt (No. 2), [1992] 1 N.Z.L.R. 676 (H.C.)) and for both intentional and unintentional defendant conduct: see, for example, McLaren Transport Ltd. v. Somerville, [1996] 3 N.Z.L.R. 424 (H.C.) (negligence); Aquaculture Corp. v. New Zealand Green Mussel Co., [1990] 3 N.Z.L.R. 299 (C.A.) (breach of confidence); Coloca v. B.P. Australia Ltd., [1992] 2 V.R. 441 (S.C.) (negligence); L. v. Robinson, [2000] 3 N.Z.L.R. 499 (H.C.) (negligence). Punitive damages for negligence causing personal injury, while rare, may be awarded if, but only if, the level of negligence is “so high that it amounts to an outrageous and flagrant disregard for the plaintiff’s safety, meriting condemnation and punishment” (McLaren Transport, supra, at p. 434, per Tipping J.; Ellison v. L., [1998] 1 N.Z.L.R. 416 (C.A.)).

57 New Zealand’s situation is somewhat complicated by its extensive statutory scheme of no-fault benefits to compensate personal injuries from accidents in lieu of compensatory damages. However, even here the New Zealand Court of Appeal has held that the legislation does not bar the award of punitive damages for the defendant’s related misconduct because the respective functions are different: the benefits under no-fault legislation compensate the victim for his or her loss without punishing the defendant, while punitive damages punish and deter the wrongdoer and are based on the quality of the defendant’s conduct (Donselaar, supra; McLaren Transport, supra; Auckland City Council v. Blundell, [1986] 1 N.Z.L.R. 732; Green v. Matheson, [1989] 3 N.Z.L.R. 564; McKenzie v. Attorney-General, [1992] 2 N.Z.L.R. 14). Punitive damages awards are a “serious and exceptional remedy” (Donselaar, supra, at p. 107). They are reserved for truly outrageous conduct where the other remedies that the defendant must bear will fall short of an adequate punishment (Dunlea v. Attorney-General, [2000] 3 N.Z.L.R. 136 (C.A.); Ellison, supra; Blundell, supra; Aquaculture Corp., supra; Cook, supra). The marking out and punishment of outrageous conduct can be achieved by a relatively modest penalty that is fairly and reasonably commensurate with the gravity of the conduct being condemned (Daniels, supra; Ellison, supra; Blundell, supra). Civil actions should be stayed until criminal actions are concluded or until it is clear that they will not be instituted (W. v. W., [1999] 2 N.Z.L.R. 1 (P.C.)). Professional disciplinary sanctions against a defendant are a “factor” but not a decisive bar against the award of punitive damages (Robinson, supra).

D. Ireland

58 In Ireland, punitive damages are said to arise from the nature of the wrong that has been committed or the manner of its commission and are intended to punish the defendant for outrageous conduct, deter the defendant and others from such conduct in the future, and “to mark the court’s particular disapproval of the defendant’s conduct in all the circumstances of the case and its decision that it should publicly be seen to have punished the defendant for such conduct” (Conway, supra, at p. 503, per Finlay C.J.). Ireland does not follow the categorical approach set out in Rookes, supra.

59 Irish courts also emphasize the need for moderation and restraint in the award of punitive damages. The Irish Supreme Court has recognized that the power to award punitive damages is a “weapon” that can be used both in defence of, and against, liberty, and that punitive damages are a “drastic, although essential, rule grounded on public policy” (Conway, supra, at p. 512, per Griffin J.; Cooper v. O’Connell, No. 85/90-96, 1997 Ireland S.C. Lexis, June 5, 1997, at p. 8, per Keane J.). As a general principle, punitive damages are not awarded if the amount of compensatory damages constitutes a sufficient public disapproval of and punishment for the particular form of the wrongdoing. See generally, Ireland, Law Reform Commission, Consultation Paper on Aggravated, Exemplary and Restitutionary Damages (1998).

E. United States

60 Punitive damages have not lacked for critics in the United States. However, in response to those critics who questioned “the propriety of this doctrine”, the Supreme Court pointed to its history: “[I]f repeated judicial decisions for more than a century are to be received as the best exposition of what the law is, the question will not admit of argument” (Day v. Woodworth, 54 U.S. (13 How.) 363 (1851), at p. 371). Nevertheless, as early as 1872, Foster J. of the New Hampshire Supreme Court surveyed the American and English precedents in Fay v. Parker, 53 N.H. 342 (1872), and, at p. 382, uttered a much-quoted condemnation calling punitive damages a “monstrous heresy” and an “unsightly and an unhealthy excrescence, deforming the symmetry of the body of the law”.

61 For practical rather than such theoretical reasons, numerous pieces of state and federal legislation now add a variety of controls, ranging from elimination of punitive damages, elimination of punitive damages in certain causes of action, caps, ratios, and diversion of portions of the money (or “kickers”) to public funds so the total amount of compensation does not go to the plaintiff. The most common form of intervention has been to cap the punitive damages at three times the compensatory award. Additionally, there are constitutional limitations (both federal and state) on the awards of punitive damages (see generally L. L. Schlueter and K. R. Redden, Punitive Damages (4th ed. 2000), vol. 1, c. 3).

62 The standard of behaviour required to trigger an award of punitive damages in the United States is usually formulated in the epithets (malicious, high-handed, oppressive, outrageous, etc.) familiar to Canadian courts.

63 In the absence of statutory intervention, the appellate courts in the United States impose a measure of discipline usually based on constitutional due process considerations. In BMW, supra, the Alabama trial jury awarded punitive damages of $4 million. The Alabama appellate court substituted an award of $2 million, which the U.S. Supreme Court quashed as so “grossly excessive” as to be unconstitutional. In its view, BMW’s conduct was not particularly reprehensible and there was an unacceptable ratio of 500:1 between the punitive damages and the actual harm. The issue of quantum was remitted to the Alabama courts and Dr. Gore’s roller coaster ride in the courts terminated in $50,000 in punitive damages: BMW of North America, Inc. v. Gore, 701 So.2d 507 (Ala. 1997). In the case of Liebeck, supra, the jury’s award of $2.7 million in punitive damages for a hot coffee burn was reduced to $480,000 by the trial judge, and the parties settled the case before appeal.

64 The United States Supreme Court has consistently rejected the suggestion of a mechanical or “mathematical formula” or “bright line” for determining whether a punitive damage award is constitutional. In the 1991 case of Pacific Mutual Life Insurance Co. v. Haslip, 499 U.S. 1 (1991), Blackmun J., for the majority, concluded that punitive damages were not per se unconstitutional and held that the punitive damages award in that case, though “close to the line” (p. 23), did not violate due process despite being a multiple of four times the compensatory damages and much exceeding the amount set by state legislation for insurance fraud. The court approved the relatively loose instructions to the jury in that case. O’Connor J. in dissent, at p. 51, objected to the lack of more specific direction. She instead proposed standards that would include factors such as (1) whether the punitive award had a reasonable relationship to the harm; (2) the defendant’s conduct, including any cover-ups, length of conduct and patterns of similar conduct; (3) the defendant’s profits; (4) the criminal sanctions against the defendant; (5) the litigation costs; (6) the defendant’s financial means; and (7) other civil actions against the defendant for the same conduct. (The reference to “(5) litigation costs” reflects the fact that in the United States even successful plaintiffs will generally have to bear their own costs.)

65 In Honda Motor Co. v. Oberg, 512 U.S. 415 (1994), the court again emphasized the need for rationality. Then in TXO Production Corp. v. Alliance Resources Corp., 509 U.S. 443 (1993), it held that even a “dramatic disparity” (p. 462) between the actual and punitive award is not a controlling factor, in part because the magnitude of the potential harm to the intended victim and the possible harm to other victims if the conduct were not deterred are important considerations. (In that case, punitive damages were about ten times the compensatory damages.) The most recent decision by the U.S. Supreme Court on punitive damage awards was handed down in May 2001 in Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 121 S.Ct. 1678 (2001). Cooper Industries held that the courts of appeal should apply a de novo, rather than abuse of discretion, standard when reviewing district court determinations of the constitutionality of punitive damages awards. The court again emphasized that, while the constitutional line is “inherently imprecise” (p. 1684) for assessing whether punishments are grossly disproportionate to the gravity of the misconduct, the three factors from BMW (1996), supra, provide “guideposts” (p. 605): (1) the degree of the defendant’s reprehensibility or culpability; (2) the relationship between the penalty and the harm or potential harm to the victim caused by the defendant’s actions; and (3) civil or criminal penalties imposed in other cases for comparable misconduct.

F. Conclusions from the Comparative Survey

66 For present purposes, I draw the following assistance from the experience in other common law jurisdictions which I believe is consistent with Canadian practice and precedent.

67 First, the attempt to limit punitive damages by “categories” does not work and was rightly rejected in Canada in Vorvis, supra, at pp. 1104-6. The control mechanism lies not in restricting the category of case but in rationally determining circumstances that warrant the addition of punishment to compensation in a civil action. It is in the nature of the remedy that punitive damages will largely be restricted to intentional torts, as in Hill, supra, or breach of fiduciary duty as in M. (K.) v. M. (H.), 1992 31 (SCC), [1992] 3 S.C.R. 6, but Vorvis itself affirmed the availability of punitive damages in the exceptional case in contract. In Denison v. Fawcett, 1958 119 (ON CA), [1958] O.R. 312, the Ontario Court of Appeal asserted in obiter that on proper facts punitive damages would be available in negligence and nuisance as well. In Robitaille v. Vancouver Hockey Club Ltd. (1981), 1981 532 (BC CA), 124 D.L.R. (3d) 228, the British Columbia Court of Appeal awarded punitive damages in a negligence case on the principle that they ought to be available whenever “the conduct of the defendant [was] such as to merit condemnation by the [c]ourt” (p. 250). This broader approach seems to be in line with most common law jurisdictions apart from England.

68 Second, there is a substantial consensus that coincides with Lord Pratt C.J.’s view in 1763 that the general objectives of punitive damages are punishment (in the sense of retribution), deterrence of the wrongdoer and others, and denunciation (or, as Cory J. put it in Hill, supra, at para. 196, they are “the means by which the jury or judge expresses its outrage at the egregious conduct”).

69 Third, there is recognition that the primary vehicle of punishment is the criminal law (and regulatory offences) and that punitive damages should be resorted to only in exceptional cases and with restraint. Where punishment has actually been imposed by a criminal court for an offence arising out of substantially the same facts, some jurisdictions, such as Australia and New Zealand, bar punitive damages in certain contexts (Gray, supra; Daniels, supra), but the dominant approach in other jurisdictions, including Canada, is to treat it as another factor, albeit a factor of potentially great importance. (Buxbaum (Litigation guardian of) v. Buxbaum, [1997] O.J. No. 5166 (QL) (C.A.); Glendale v. Drozdzik (1993), 1993 6867 (BC CA), 77 B.C.L.R. (2d) 106 (C.A.); Pollard v. Gibson (1986), 1 Y.R. 167 (S.C.); Joanisse v. Y. (D.) (1995), 1995 2783 (BC SC), 15 B.C.L.R. (3d) 224 (S.C.); Canada v. Lukasik (1985), 1985 1146 (AB QB), 18 D.L.R. (4th) 245 (Alta. Q.B.); Wittig v. Wittig (1986), 1986 3081 (SK QB), 53 Sask. R. 138 (Q.B.)) The Ontario Law Reform Commission, supra, recommended that the “court should be entitled to consider the fact and adequacy of any prior penalty imposed in any criminal or other similar proceeding brought against the defendant” (p. 46).

70 Fourth, the incantation of the time-honoured pejoratives (“high-handed”, “oppressive”, “vindictive”, etc.) provides insufficient guidance (or discipline) to the judge or jury setting the amount. Lord Diplock in Cassell, supra, at p. 1129, called these the “whole gamut of dyslogistic judicial epithets”. A more principled and less exhortatory approach is desirable.

71 Fifth, all jurisdictions seek to promote rationality. In directing itself to the punitive damages, the court should relate the facts of the particular case to the underlying purposes of punitive damages and ask itself how, in particular, an award would further one or other of the objectives of the law, and what is the lowest award that would serve the purpose, i.e., because any higher award would be irrational.

72 Sixth, it is rational to use punitive damages to relieve a wrongdoer of its profit where compensatory damages would amount to nothing more than a licence fee to earn greater profits through outrageous disregard of the legal or equitable rights of others.

73 Seventh, none of the common law jurisdictions has adopted (except by statute) a formulaic approach, as advocated by the intervener the Insurance Council of Canada in this appeal, such as a fixed cap or fixed ratio between compensatory and punitive damages. The proper focus is not on the plaintiff’s loss but on the defendant’s misconduct. A mechanical or formulaic approach does not allow sufficiently for the many variables that ought to be taken into account in arriving at a just award.

74 Eighth, the governing rule for quantum is proportionality. The overall award, that is to say compensatory damages plus punitive damages plus any other punishment related to the same misconduct, should be rationally related to the objectives for which the punitive damages are awarded (retribution, deterrence and denunciation). Thus there is broad support for the “if, but only if” test formulated, as mentioned, in Rookes, supra, and affirmed here in Hill, supra.

75 Ninth, it has become evident that juries can and should receive more guidance and help from the judges in terms of their mandate. They should be told in some detail about the function of punitive damages and the factors that govern both the award and the assessment of a proper amount. Juries should not be thrown into their assignment without any help, then afterwards be criticized for the result.

76 Tenth, and finally, there is substantial consensus (even the United States is moving in this direction) that punitive damages are not at large (as pointed out by Cory J. in Hill, supra, at para. 197) and that an appellate court is entitled to intervene if the award exceeds the outer boundaries of a rational and measured response to the facts of the case.

77 With the benefit of these general principles, I now turn to the specific issues raised by this appeal.

(1) Punitive Damages for Breach of Contract

78 This, as noted, is a breach of contract case. In Vorvis, supra, this Court held that punitive damages are recoverable in such cases provided the defendant’s conduct said to give rise to the claim is itself “an actionable wrong” (p. 1106). The scope to be given this expression is the threshold question in this case, i.e., is a breach of an insurer’s duty to act in good faith an actionable wrong independent of the loss claim under the fire insurance policy? Vorvis itself was a case about the employer’s breach of an employment contract. This is how McIntyre J. framed the rule at pp. 1105-6:

When then can punitive damages be awarded? It must never be forgotten that when awarded by a judge or jury, a punishment is imposed upon a person by a Court by the operation of the judicial process. What is it that is punished? It surely cannot be merely conduct of which the Court disapproves, however strongly the judge may feel. Punishment may not be imposed in a civilized community without a justification in law. The only basis for the imposition of such punishment must be a finding of the commission of an actionable wrong which caused the injury complained of by the plaintiff. [Emphasis added.]

This view, McIntyre J. said (at p. 1106), “has found approval in the Restatement on the Law of Contracts 2d in the United States”, which reads as follows:

Punitive damages are not recoverable for a breach of contract unless the conduct constituting the breach is also a tort for which punitive damages are recoverable. [Emphasis added.]

Applying these principles in Vorvis, McIntyre J. stated, at p. 1109:

Each party had the right to terminate the contract without the consent of the other, and where the employment contract was terminated by the employer, the appellant was entitled to reasonable notice of such termination or payment of salary and benefits for the period of reasonable notice. The termination of the contract on this basis by the employer is not a wrong in law and, where the reasonable notice is given or payment in lieu thereof is made, the plaintiff — subject to a consideration of aggravated damages which have been allowed in some cases but which were denied in this case — is entitled to no further remedy . . . . [Emphasis added.]

Wilson J., with whom L’Heureux-Dubé J. concurred, dissented. She did not agree “that punitive damages can only be awarded when the misconduct is in itself an ‘actionable wrong’”. She stated, at p. 1130:

In my view, the correct approach is to assess the conduct in the context of all the circumstances and determine whether it is deserving of punishment because of its shockingly harsh, vindictive, reprehensible or malicious nature. Undoubtedly some conduct found to be deserving of punishment will constitute an actionable wrong but other conduct might not.

79 In the case at bar, Pilot acknowledges that an insurer is under a duty of good faith and fair dealing. Pilot says that this is a contractual duty. Vorvis, it says, requires a tort. However, in my view, a breach of the contractual duty of good faith is independent of and in addition to the breach of contractual duty to pay the loss. It constitutes an “actionable wrong” within the Vorvis rule, which does not require an independent tort. I say this for several reasons.

80 First, McIntyre J. chose to use the expression “actionable wrong” instead of “tort” even though he had just reproduced an extract from the Restatement which does use the word tort. It cannot be an accident that McIntyre J. chose to employ a much broader expression when formulating the Canadian test.

81 Second, in Royal Bank of Canada v. W. Got & Associates Electric Ltd., 1999 714 (SCC), [1999] 3 S.C.R. 408, at para. 26, this Court, referring to McIntyre J.’s holding in Vorvis, said “the circumstances that would justify punitive damages for breach of contract in the absence of actions also constituting a tort are rare” (emphasis added). Rare they may be, but the clear message is that such cases do exist. The Court has thus confirmed that punitive damages can be awarded in the absence of an accompanying tort.

82 Third, the requirement of an independent tort would unnecessarily complicate the pleadings, without in most cases adding anything of substance. Central Trust Co. v. Rafuse, 1986 29 (SCC), [1986] 2 S.C.R. 147, held that a common law duty of care sufficient to found an action in tort can arise within a contractual relationship, and in that case proceeded with the analysis in tort instead of contract to deprive an allegedly negligent solicitor of the benefit of a limitation defence. To require a plaintiff to formulate a tort in a case such as the present is pure formalism. An independent actionable wrong is required, but it can be found in breach of a distinct and separate contractual provision or other duty such as a fiduciary obligation.

83 I should add that insurance companies have also asserted claims for punitive damages against their insured for breach of the mutual “good faith” obligation in insurance contracts. In Andrusiw v. Aetna Life Insurance Co. of Canada (2001), 2001 61004 (AB QB), 289 A.R. 1 (Q.B.), the court awarded $20,000 in punitive damages against an Aetna policy holder in addition to an order for the repayment of $260,000 in disability payments. The insurance company was not required to identify a separate tort to ground its claim for punitive damages. In that case it was the misconduct of the policy holder, not the insurance company, that was seen as such a marked departure from ordinary standards of decent behaviour as to invite the censure of punitive damages, per Murray J. at paras. 84-85:

This leaves the question of whether or not the plaintiff’s conduct was so reprehensible and high-handed that he should be punished for his behaviour. Counsel for the defendant makes the point that the plaintiff embarked on a deliberate course of conduct to misrepresent facts to the defendant in order to continue to collect disability benefits. If the only consequence of this behaviour is forfeiture of his claim then in effect he is no worse off than if he had been truthful in the first place and deterrence which is one of the objects of granting punitive damages is given no effect.

A great deal has been made in the case law, to which this court was referred, of the fact that insurers vis-à-vis their insureds are in a superior bargaining position and one which places the insureds in positions of dependency and vulnerability. Equally, insurers must not be looked upon as fair game. It is a two-way street founded upon the principle of utmost good faith arising from the very nature of the contract. Thus, it is appropriate that punitive damages be awarded and I do so in the sum of $20,000.00.

I refrain from any comment on the correctness of this award, but to those who subscribe to “the sting” approach to punitive damages, I pose the question whether an award of $20,000 against a cheating policy holder in the Aetna case has at least as much “sting”, or possibly more, than the award of $1 million against Pilot in this case.

(2) Was the Claim for Punitive Damages Properly Pleaded?

84 The respondent says that even if a separate claim arising under the insurance contract could provide the basis for punitive damages, none was pleaded in this case.

85 In other words, while “punitive and exemplary damages” are explicitly requested in para. 13 of the statement of claim, the material facts necessary for the grant of such an award are not spelled out in the body of the pleading. Further, the respondent in its cross-appeal says that even if the plaintiff has established an “independent actionable wrong”, she failed to prove any separate and distinct damage flowing from it. The appellant thus failed, Pilot says, to meet the Vorvis requirements and her claim for punitive damages ought to have been dismissed.

86 There is some case law that says a claim for punitive damages need not be specifically pleaded as it is included conceptually in a claim for general damages: Edwards v. Harris-Intertype (Canada) Ltd. (1983), 1983 1611 (ON SC), 40 O.R. (2d) 558 (H.C.), aff’d (1984), 1984 2070 (ON CA), 9 D.L.R. (4th) 319 (Ont. C.A.); Grenn v. Brampton Poultry Co. (1959), 1959 394 (ON CA), 18 D.L.R. (2d) 9 (Ont. C.A.), Starkman v. Delhi Court Ltd. (1960), 1960 413 (ON SC), 24 D.L.R. (2d) 152 (Ont. H.C.), aff’d (1961), 1961 195 (ON CA), 28 D.L.R. (2d) 269 (Ont. C.A.); Gastebled v. Stuyck (1973), 1973 2285 (FC), 12 C.P.R. (2d) 102 (F.C.T.D.), aff’d (1974), 1974 2447 (FCA), 15 C.P.R. (2d) 137 (F.C.A.); Paragon Properties Ltd. v. Magna Envestments Ltd. (1972), 1972 ALTASCAD 8, 24 D.L.R. (3d) 156 (Alta. S.C., App. Div.). In my view, the suggestion that no pleading is necessary overlooks the basic proposition in our justice system that before someone is punished they ought to have advance notice of the charge sufficient to allow them to consider the scope of their jeopardy as well as the opportunity to respond to it. This can only be assured if the claim for punitive damages, as opposed to compensatory damages, is not buried in a general reference to general damages. This principle, which is really no more than a rule of fairness, is made explicit in the civil rules of some of our trial courts. For example, in Saskatchewan the Queen’s Bench Rules require that claims for punitive damages be expressly pleaded and specify the misconduct which is claimed to give rise to such damages (Rieger v. Burgess, 1988 209 (SK CA), [1988] 4 W.W.R. 577 (Sask. C.A.); Lauscher v. Berryere (1999), 1999 12242 (SK CA), 172 D.L.R. (4th) 439 (Sask. C.A.)). Rule 25.06(9) of the Ontario Rules of Civile Procedure also has the effect of requiring that punitive damages claims be expressly pleaded. It is quite usual, of course, for the complexion of a case to evolve over time, but a pleading can always be amended on terms during the proceedings, depending on the existence and extent of prejudice not compensable in costs, and the justice of the case.

87 One of the purposes of a statement of claim is to alert the defendant to the case it has to meet, and if at the end of the day the defendant is surprised by an award against it that is a multiple of what it thought was the amount in issue, there is an obvious unfairness. Moreover, the facts said to justify punitive damages should be pleaded with some particularity. The time-honoured adjectives describing conduct as “harsh, vindictive, reprehensible and malicious” (per McIntyre J. in Vorvis, supra, p. 1108) or their pejorative equivalent, however apt to capture the essence of the remedy, are conclusory rather than explanatory.

88 Whether or not a defendant has in fact been taken by surprise by a weak or defective pleading will have to be decided in the circumstances of a particular case.

89 In this case, the plaintiff specifically asked for punitive damages in her statement of claim and if the respondent was in any doubt about the facts giving rise to the claim, it ought to have applied for particulars and, in my opinion, it would have been entitled to them.

90 However, the respondent did not apply for particulars, and I think there is sufficient detail in the statement of claim to show that its failure to do so was not a self-inflicted injustice. There was no surprise except perhaps as to the quantum, which resulted in an amendment of the statement of claim at trial. Quite apart from the advance notice that she was seeking punitive damages (para. 1(e)), the appellant specifically pleaded the basis for the independent “actionable wrong” in para. 10:

10. The Plaintiff pleads an implied term of the insurance contract was a covenant of good faith and fair dealings which required the Defendant, Pilot Insurance Company to deal fairly and in good faith in handling the claim of the Plaintiff.

91 The appellant also pleaded that Pilot’s manner of dealing with her claim had created “hardship” of which “the Defendants, through their agents and employees always had direct and ongoing knowledge” (para. 8). In para. 14 she pleaded that “as a result of the actions of the Defendants, the Plaintiff has suffered and continues to suffer great emotional stress” (although there was no claim for aggravated damages). The respondent specifically denied acting in bad faith (Statement of Defence and Counterclaim of the Defendant, at para. 6). The statement of claim was somewhat deficient in failing to relate the plea for punitive damages to the precise facts said to give rise to the outrage, but Pilot was content to go to trial on this pleading and I do not think it should be heard to complain about it at this late date.

92 As to the respondent’s objection that the pleading does not allege separate and distinct damages flowing from the independent actionable wrong, the respondent’s argument overlooks the fact that punitive damages are directed to the quality of the defendant’s conduct, not the quantity (if any) of the plaintiff’s loss. As Cory J. observed in Hill, supra, at para. 196, “[p]unitive damages bear no relation to what the plaintiff should receive by way of compensation. Their aim is not to compensate the plaintiff, but rather to punish the defendant. It is the means by which the jury or judge expresses its outrage at the egregious conduct of the defendant”. In any event, there is a good deal of evidence of emotional stress and financial cost over and above the loss that would have been incurred had the claim been settled in good faith within a reasonable time.

(3) Was the Jury Charge Adequate?

93 The respondent argues that the trial judge did not give the jury adequate guidance on how to assess punitive damages. There is considerable merit in this submission. The judge’s charge on this point was skeletal. It is my view, for the reasons already discussed, that the charge on punitive damages should not be given almost as an afterthought but should be understood as an important source of control and discipline. The jurors should not be left to guess what their role and function is.

94 To this end, not only should the pleadings of punitive damages be more rigorous in the future than in the past (see para. 87 above), but it would be helpful if the trial judge’s charge to the jury included words to convey an understanding of the following points, even at the risk of some repetition for emphasis. (1) Punitive damages are very much the exception rather than the rule, (2) imposed only if there has been high-handed, malicious, arbitrary or highly reprehensible misconduct that departs to a marked degree from ordinary standards of decent behaviour. (3) Where they are awarded, punitive damages should be assessed in an amount reasonably proportionate to such factors as the harm caused, the degree of the misconduct, the relative vulnerability of the plaintiff and any advantage or profit gained by the defendant, (4) having regard to any other fines or penalties suffered by the defendant for the misconduct in question. (5) Punitive damages are generally given only where the misconduct would otherwise be unpunished or where other penalties are or are likely to be inadequate to achieve the objectives of retribution, deterrence and denunciation. (6) Their purpose is not to compensate the plaintiff, but (7) to give a defendant his or her just desert (retribution), to deter the defendant and others from similar misconduct in the future (deterrence), and to mark the community’s collective condemnation (denunciation) of what has happened. (8) Punitive damages are awarded only where compensatory damages, which to some extent are punitive, are insufficient to accomplish these objectives, and (9) they are given in an amount that is no greater than necessary to rationally accomplish their purpose. (10) While normally the state would be the recipient of any fine or penalty for misconduct, the plaintiff will keep punitive damages as a “windfall” in addition to compensatory damages. (11) Judges and juries in our system have usually found that moderate awards of punitive damages, which inevitably carry a stigma in the broader community, are generally sufficient.

95 These particular expressions are not, of course, obligatory. What is essential in a particular case will be a function of its particular circumstances, the need to emphasize the nature, scope and exceptional nature of the remedy, and fairness to both sides.

96 The trial judge should keep in mind that the standard of appellate review applicable to punitive damages ultimately awarded, is that a reasonable jury, properly instructed, could have concluded that an award in that amount, and no less, was rationally required to punish the defendant’s misconduct, as discussed below.

97 If counsel can agree on a “bracket” or “range” of an appropriate award, the trial judge should convey these figures to the jury, but at the present time specific figures should not be mentioned in the absence of such agreement (Hill, supra, per Cory J., at paras. 162-63). (This prohibition may have to be reexamined in future, based on further experience.) Counsel should also consider the desirability of asking the trial judge to advise the jury of awards of punitive damages made in comparable circumstances that have been sustained on appeal.

98 The foregoing suggestions are put forward in an effort to be helpful rather than dogmatic. They grow out of the observation in Hill that punitive damages are not “at large” (para. 197). Unless punitive damages can be approached rationally they ought not to be awarded at all. To the extent these suggestions are considered useful, they will obviously have to be both modified and elaborated to assist the jury on the facts of a particular case. The point, simply, is that jurors should not be left in any doubt about what they are to do and how they are to go about it.

99 It is evident that I am suggesting a more ample charge on the issue of punitive damages than was given in this case. Finlayson J.A. said that he was “not entirely happy with the trial judge’s charge to the jury on the issue of punitive damages” (p. 661), and Laskin J.A. agreed that “[t]he trial judge might have given the jury more help than he did” (p. 656). However, both Finlayson and Laskin JJ.A. agreed that the jury charge covered the essentials, however lightly. This conclusion is reinforced by the fact that no objection was made by either counsel. With some hesitation, I agree with the Court of Appeal, unanimous on this point, that in the circumstances this ground of appeal should be rejected.

(4) Reviewing the Jury Award

(a) Whether the Award of Punitive Damages in This Case was a Rational Response to the Respondent’s Misconduct

100 The applicable standard of review for “rationality” was articulated by Cory J. in Hill, supra, at para. 197:

Unlike compensatory damages, punitive damages are not at large. Consequently, courts have a much greater scope and discretion on appeal. The appellate review should be based upon the court’s estimation as to whether the punitive damages serve a rational purpose. In other words, was the misconduct of the defendant so outrageous that punitive damages were rationally required to act as deterrence?

101 The “rationality” test applies both to the question of whether an award of punitive damages should be made at all, as well as to the question of its quantum.

102 The respondent claims that an insurer is entirely within its rights to thoroughly investigate a claim and exercise caution in evaluating the circumstances. It is not required to accept the initial views of its investigators. It is perfectly entitled to pursue further inquiries. I agree with these points. The problem here is that Pilot embarked on a “train of thought” as early as February 25, 1994 (see para. 7 above) that led to the arson trial, with nothing to go on except the fact that its policy holder had money problems.

103 The “train of thought” mentioned in the letter to Pilot from Derek Francis kept going long after the requirements of due diligence or prudent practice had been exhausted. There is a difference between due diligence and wilful tunnel vision. The jury obviously considered this case to be an outrageous example of the latter. In my view, an award of punitive damages (leaving aside the issue of quantum for the moment) was a rational response on the jury’s part to the evidence. It was not an inevitable or unavoidable response, but it was a rational response to what the jury had seen and heard. The jury was obviously incensed at the idea that the respondent would get away with paying no more than it ought to have paid after its initial investigation in 1994 (plus costs). It obviously felt that something more was required to demonstrate to Pilot that its bad faith dealing with this loss claim was not a wise or profitable course of action. The award answered a perceived need for retribution, denunciation and deterrence.

104 The intervener, the Insurance Council of Canada, argues that the award of punitive damages will over-deter insurers from reviewing claims with due diligence, thus lead to the payment of unmeritorious claims, and in the end drive up insurance premiums. This would only be true if the respondent’s treatment of the appellant is not an isolated case but is widespread in the industry. If, as I prefer to believe, insurers generally take seriously their duty to act in good faith, it will only be rogue insurers or rogue files that will incur such a financial penalty, and the extra economic cost inflicted by punitive damages will either cause the delinquents to mend their ways or, ultimately, move them on to lines of work that do not call for a good faith standard of behaviour.

105 The Ontario Court of Appeal was unanimous that punitive damages in some amount were justified and I agree with that conclusion. This was an exceptional case that justified an exceptional remedy. The respondent’s cross-appeal will therefore be dismissed.

106 We now come to the issue of quantum.

(b) Whether the Jury’s Award of $1 Million in Punitive Damages Should Be Restored

107 In Hill, supra, Cory J., while emphasizing the overriding obligation of rationality, also recognized that the jury must be given some leeway to do its job. The issue of punitive damages, after all, is a matter that has been confided in the first instance to their discretion. Thus, to be reversed, their award of punitive damages must be “so inordinately large as obviously to exceed the maximum limit of a reasonable range within which the jury may properly operate” (para. 159). Putting these two notions together, the test is whether a reasonable jury, properly instructed, could have concluded that an award in that amount, and no less, was rationally required to punish the defendant’s misconduct.

108 This test provides an appellate court with supervisory powers over punitive damages that are more interventionist than in the case of other jury awards of general damages, where the courts may only intervene if the award is “so exorbitant or so grossly out of proportion [to the injury] as to shock the court’s conscience and sense of justice” (Hill, supra, at para. 159; Walker v. CFTO Ltd. (1987), 1987 126 (ON CA), 59 O.R. (2d) 104 (C.A.)). In the case of punitive damages, the emphasis is on the appellate court’s obligation to ensure that the award is the product of reason and rationality. The focus is on whether the court’s sense of reason is offended rather than on whether its conscience is shocked.

109 If the award of punitive damages, when added to the compensatory damages, produces a total sum that is so “inordinately large” that it exceeds what is “rationally” required to punish the defendant, it will be reduced or set aside on appeal.

110 An award that is higher than required to fulfil its purpose is, by definition, irrational. The more difficult task is to determine what is “inordinate”. Here, I think, the Court must come to grips with the issue of proportionality.

111 I earlier referred to proportionality as the key to the permissible quantum of punitive damages. Retribution, denunciation and deterrence are the recognized justification for punitive damages, and the means must be rationally proportionate to the end sought to be achieved. A disproportionate award overshoots its purpose and becomes irrational. A less than proportionate award fails to achieve its purpose. Thus a proper award must look at proportionality in several dimensions, including:

(i) Proportionate to the Blameworthiness of the Defendant’s Conduct

112 The more reprehensible the conduct, the higher the rational limits to the potential award. The need for denunciation is aggravated where, as in this case, the conduct is persisted in over a lengthy period of time (two years to trial) without any rational justification, and despite the defendant’s awareness of the hardship it knew it was inflicting (indeed, the respondent anticipated that the greater the hardship to the appellant, the lower the settlement she would ultimately be forced to accept).

113 The level of blameworthiness may be influenced by many factors, but some of the factors noted in a selection of Canadian cases include:

(1) whether the misconduct was planned and deliberate: Patenaude v. Roy (1994), 1994 6107 (QC CA), 123 D.L.R. (4th) 78 (Que. C.A.), at p. 91;

(2) the intent and motive of the defendant: Recovery Production Equipment Ltd. v. McKinney Machine Co. (1998), 1998 ABCA 239, 223 A.R. 24 (C.A.), at para. 77;

(3) whether the defendant persisted in the outrageous conduct over a lengthy period of time: Mustaji v. Tjin (1996), 1996 1907 (BC CA), 30 C.C.L.T. (2d) 53 (B.C.C.A.), Québec (Curateur public) v. Syndicat national des employés de l’Hôpital St-Ferdinand (1994), 1994 6112 (QC CA), 66 Q.A.C. 1, Matusiak v. British Columbia and Yukon Territory Building and Construction Trades Council, [1999] B.C.J. No. 2416 (QL) (S.C.);

(4) whether the defendant concealed or attempted to cover up its misconduct: Gerula v. Flores (1995), 1995 1096 (ON CA), 126 D.L.R. (4th) 506 (Ont. C.A.), at p. 525, Walker v. D’Arcy Moving & Storage Ltd. (1999), 1999 2808 (ON CA), 117 O.A.C. 367 (C.A.), United Services Funds (Trustees) v. Hennessey, [1994] O.J. No. 1391 (QL) (Gen. Div.), at para. 58;

(5) the defendant’s awareness that what he or she was doing was wrong: Williams v. Motorola Ltd. (1998), 1998 5023 (ON CA), 38 C.C.E.L. (2d) 76 (Ont. C.A.), and Procor Ltd. v. U.S.W.A. (1990), 1990 6637 (ON SC), 71 O.R. (2d) 410 (H.C.), at p. 433;

(6) whether the defendant profited from its misconduct: Claiborne Industries Ltd. v. National Bank of Canada (1989), 1989 183 (ON CA), 69 O.R. (2d) 65 (C.A.);

(7) whether the interest violated by the misconduct was known to be deeply personal to the plaintiff (e.g., professional reputation (Hill, supra)) or a thing that was irreplaceable (e.g., the mature trees cut down by the real estate developer in Horseshoe Bay Retirement Society v. S.I.F. Development Corp. (1990), 1990 8047 (BC SC), 66 D.L.R. (4th) 42 (B.C.S.C.)); see also Kates v. Hall (1991), 1991 1127 (BC CA), 53 B.C.L.R. (2d) 322 (C.A.). Special interests have included the reproductive capacity of the plaintiff deliberately sterilized by an irreversible surgical procedure while the plaintiff was confined in a provincial mental institution, although no award of punitive damages was made on the facts (Muir v. Alberta, 1996 7287 (AB QB), [1996] 4 W.W.R. 177 (Alta. Q.B.)); the deliberate publication of an informant’s identity (R. (L.) v. Nyp (1995), 25 C.C.L.T. (2d) 309 (Ont. Ct. (Gen. Div.)). In Weinstein v. Bucar, 1990 11072 (MB QB), [1990] 6 W.W.R. 615 (Man. Q.B.), the defendant shot and killed plaintiffs’ three companion and breeding German Shepherds who had merely wandered onto the defendant’s property from a neighbouring yard. Here the “property” was sentimental, not replaceable, and, unlike the trees, themselves sentient beings.

(ii) Proportionate to the Degree of Vulnerability of the Plaintiff

114 The financial or other vulnerability of the plaintiff, and the consequent abuse of power by a defendant, is highly relevant where there is a power imbalance. In Norberg v. Wynrib, 1992 65 (SCC), [1992] 2 S.C.R. 226, for example, speaking of a physician who had used his access to drugs to purchase sex from a female patient, McLachlin J. (as she then was) stated, at p. 276:

Society has an abiding interest in ensuring that the power entrusted to physicians by us, both collectively and individually, not be used in corrupt ways . . . .

A similar point was made by Laskin J.A. in the present case (at p. 659):

[V]indicating the goal of deterrence is especially important in first party insurance cases. Insurers annually deal with thousands and thousands of claims by their insureds. A significant award was needed to deter Pilot and other insurers from exploiting the vulnerability of insureds, who are entirely dependent on their insurers when disaster strikes.

115 I add two cautionary notes on the issue of vulnerability. First, this factor militates against the award of punitive damages in most commercial situations, particularly where the cause of action is contractual and the problem for the court is to sort out the bargain the parties have made. Most participants enter the marketplace knowing it is fuelled by the aggressive pursuit of self-interest. Here, on the other hand, we are dealing with a homeowner’s “peace of mind” contract.

116 Second, it must be kept in mind that punitive damages are not compensatory. Thus the appellant’s pleading of emotional distress in this case is only relevant insofar as it helps to assess the oppressive character of the respondent’s conduct. Aggravated damages are the proper vehicle to take into account the additional harm caused to the plaintiff’s feelings by reprehensible or outrageous conduct on the part of the defendant. Otherwise there is a danger of “double recovery” for the plaintiff’s emotional stress, once under the heading of compensation and secondly under the heading of punishment.

(iii) Proportionate to the Harm or Potential Harm Directed Specifically at the Plaintiff

117 The jury is not a general ombudsman or roving Royal Commission. There is a limited role for the plaintiff as private attorney general. It would be irrational to provide the plaintiff with an excessive windfall arising out of a defendant’s scam of which the plaintiff was but a minor or peripheral victim. On the other hand, malicious and high-handed conduct which could be expected to cause severe injury to the plaintiff is not necessarily excused because fortuitously it results in little damage.

(iv) Proportionate to the Need for Deterrence

118 The theory is that it takes a large whack to wake up a wealthy and powerful defendant to its responsibilities. The appellant’s argument is that the punitive damages award of $1 million represents less than one half of one percent of Pilot’s net worth. This is a factor, but it is a factor of limited importance.

119 A defendant’s financial power may become relevant (1) if the defendant chooses to argue financial hardship, or (2) it is directly relevant to the defendant’s misconduct (e.g., financial power is what enabled the defendant Church of Scientology to sustain such an outrageous campaign for so long against the plaintiff in Hill, supra), or (3) other circumstances where it may rationally be concluded that a lesser award against a moneyed defendant would fail to achieve deterrence.

120 Deterrence is an important justification for punitive damages. It would play an even greater role in this case if there had been evidence that what happened on this file were typical of Pilot’s conduct towards policyholders. There was no such evidence. The deterrence factor is still important, however, because the egregious misconduct of middle management was known at the time to top management, who took no corrective action.

121 The fact the respondent’s assets of $231 million were mentioned to the jury in this case was unhelpful. Pilot was obviously a substantial corporation. Disclosure of detailed financial information before liability is established may wrongly influence the jury to find liability where none exists (i.e., the subliminal message may be “What’s a $345,000 insurance claim to a $231 million company?”). Moreover, pre-trial discovery of financial capacity would unnecessarily prolong the pre-trial proceedings and prematurely switch the focus from the plaintiff’s claim for compensation to the defendant’s capacity to absorb punishment. In any event, the court should hesitate to attribute anthropomorphic qualities to large corporations (i.e., the punishment should “sting”).

122 Where a trial judge is concerned that the claim for punitive damages may affect the fairness of the liability trial, bifurcated proceedings may be appropriate. On the facts of this case, no harm was done by the procedure followed, including the mention of the $231 million figure.

(v) Proportionate, Even After Taking Into Account the Other Penalties, Both Civil and Criminal, Which Have Been or Are Likely to Be Inflicted on the Defendant for the Same Misconduct

123 Compensatory damages also punish. In many cases they will be all the “punishment” required. To the extent a defendant has suffered other retribution, denunciation or deterrence, either civil or criminal, for the misconduct in question, the need for additional punishment in the case before the court is lessened and may be eliminated. In Canada, unlike some other common law jurisdictions, such “other” punishment is relevant but it is not necessarily a bar to the award of punitive damages. The prescribed fine, for example, may be disproportionately small to the level of outrage the jury wishes to express. The misconduct in question may be broader than the misconduct proven in evidence in the criminal or regulatory proceeding. The legislative judgment fixing the amount of the potential fine may be based on policy considerations other than pure punishment. The key point is that punitive damages are awarded “if, but only if” all other penalties have been taken into account and found to be inadequate to accomplish the objectives of retribution, deterrence, and denunciation. The intervener, the Insurance Council of Canada, argues that the discipline of insurance companies should be left to the regulator. Nothing in the appeal record indicates that the Registrar of Insurance (now the Superintendent of Financial Services) took an interest in this case prior to the jury’s unexpectedly high award of punitive damages.

(vi) Proportionate to the Advantage Wrongfully Gained by a Defendant from the Misconduct

124 A traditional function of punitive damages is to ensure that the defendant does not treat compensatory damages merely as a licence to get its way irrespective of the legal or other rights of the plaintiff. Thus in Horseshoe Bay Retirement Society, supra, a real estate developer cut down mature trees on the plaintiff’s property to improve the view from neighbouring lots which it was developing for sale. The defendant appeared to have calculated that enhanced prices for its properties would exceed any “compensation” that it might be required to pay to the plaintiff. Punitive damages of $100,000 were awarded to reduce the profits and deter “like-minded” developers (p. 50). For a similar case, see Nantel v. Parisien (1981), 18 C.C.L.T. 79 (Ont. H.C.), per Galligan J., at p. 87, “. . . the law would say to the rich and powerful, ‘Do what you like, you will only have to make good the plaintiff’s actual financial loss, which compared to your budget is negligible’”. In Claiborne Industries, supra, an award of punitive damages was made against the defendant bank in an amount sufficient to ensure that it did not profit from its outrageous conduct (p. 106).

125 On the other hand, care must be taken not to employ the “wrongful profit” factor irrationally. Thus, in Lubrizol Corp. v. Imperial Oil Ltd. (1994), 1994 19680 (FC), 84 F.T.R. 197, the court ordered the defendant to account to the plaintiff for all profits gained by infringing the plaintiff’s patent, with interest, then added $15 million in punitive damages (without waiting for the profits to be ascertained) because, per Cullen J., “[t]he volume of [patented] product sold, although not quantified, must be enormous” and the defendant was “a large corporation with annual sales of 10 billion dollars” (p. 209). The duplicative remedies thus relieved the defendant of the profit twice, once through the accounting remedy and a second time (at least in part) through an award of punitive damages. The trial judge’s approach was reversed on appeal ((1996), 1996 4042 (FCA), [1996] 3 F.C. 40).

126 In the present case, the effort to force the appellant into a disadvantageous settlement having failed, it is not alleged that the respondent profited from its misconduct.

(5) The Usefulness of Ratios

127 The respondent and its supporting intervener suggest that an award of $1 million in punitive damages is out of line because compensatory damages were ultimately assessed only at about $345,000. The result, they argue, is an improper ratio. It is apparent from what has already been said, however, that proportionality is a much broader concept than the simple relationship between punitive damages and compensatory damages. That relationship, moreover, is not even the most relevant because it puts the focus on the plaintiff’s loss rather than where it should be, on the defendant’s misconduct. If a ratio is to be used what should the ratio measure? The fact that compensatory damages are quantified in dollars and cents is temptingly useful, but wholly inadequate, for example, in a case where outrageous misconduct has fortuitously (and fortunately) resulted in a small financial loss. Potential, as well as actual, harm is a reasonable measure of misconduct, and so are the other factors, already mentioned, such as motive, planning, vulnerability, abuse of dominance, other fines or penalties, and so on. None of these features are captured by the ratio of punitive damages to compensatory damages. Adoption of such a ratio, while easy to supervise, would do a disservice to the unavoidable complexity of the analysis. It would in fact undermine the nuanced principles on which the concept of punitive damages has been justified. There is no doubt at all that evaluation of outrageous conduct in terms of dollars and cents is a difficult and imprecise task, but so is evaluating the worth of a cracked skull, a lost business opportunity or a shattered reputation. Yet all these things are done every day in the courts in the calculation of compensatory damages without resort to formulae or arbitrary rules such as ratios.

(6) Conclusion on “Rationality”

128 I would not have awarded $1 million in punitive damages in this case but in my judgment the award is within the rational limits within which a jury must be allowed to operate. The award was not so disproportionate as to exceed the bounds of rationality. It did not overshoot its purpose. I have already outlined the reasons why I believe this to be the case.

129 The jury followed the “if, but only if” model, i.e., punitive damages should be awarded “if, but only if” the compensatory award is insufficient. The form and order of the questions put to the jury required them first of all to deal with compensation for the loss of the plaintiff’s house (replacement or cash value), its contents, and any increase in her living and moving expenses. Only after those matters had been dealt with was the jury instructed to turn their minds to a final question on punitive damages. They were clearly aware that compensatory damages might well be sufficient punishment to avoid a repetition of the offence and a deterrent to others. In this case, the jury obviously concluded that the compensatory damages ($345,000) were not sufficient for those purposes. It was no more than the respondent had contractually obligated itself to pay under the insurance policy. In this case, the power imbalance was highly relevant. Pilot holds itself out to the public as a sure guide to a “safe harbour”. In its advertising material it refers to itself as “Your Pilot” and makes such statements as:

At Pilot Insurance Company, guiding people like you into safe harbours has been our mission for nearly 75 years.

Insurance contracts, as Pilot’s self-description shows, are sold by the insurance industry and purchased by members of the public for peace of mind. The more devastating the loss, the more the insured may be at the financial mercy of the insurer, and the more difficult it may be to challenge a wrongful refusal to pay the claim. Deterrence is required. The obligation of good faith dealing means that the appellant’s peace of mind should have been Pilot’s objective, and her vulnerability ought not to have been aggravated as a negotiating tactic. It is this relationship of reliance and vulnerability that was outrageously exploited by Pilot in this case. The jury, it appears, decided a powerful message of retribution, deterrence and denunciation had to be sent to the respondent and they sent it.

130 The respondent points out that there is no evidence this case represents a deliberate corporate strategy as opposed to an isolated, mishandled file that ran amok. This is true, but it is also true that Pilot declined to call evidence to explain why this file ran amok, and what steps, if any, have been taken to prevent a recurrence.

131 The respondent also argues that at the end of the day, it did not profit financially from its misbehaviour. This may also be true, but if so, that result was not for want of trying. The respondent clearly hoped to starve the appellant into a cheap settlement. Crabbe’s letter of June 9, 1994, quoted earlier, suggests as much. That it failed to do so is due in no small part to appellant’s counsel who took a hotly contested claim into an eight-week jury trial on behalf of a client who was effectively without resources of her own; and who obviously could have been starved into submission but for his firm’s intervention on her behalf.

132 While, as stated, I do not consider the “ratio” test to be an appropriate indicator of rationality, the ratio of punitive damages to compensatory damages in the present case would be either a multiple of three (if only the insurance claim of $345,000 is considered) or a multiple of less than two (if the claim plus the award of solicitor‑client costs is thought to be the total compensation). Either way, the ratio is well within what has been considered “rational” in decided cases.

133 The majority opinion of the Ontario Court of Appeal recognized that punitive damages are not “at large” and appellate courts have “much greater scope and discretion on appeal” than they do in the case of general damages (Hill, supra, at para. 197). If the court considers the award or its quantum to be irrational, it is its duty to interfere.

134 This was the view taken by the majority judgment of Finlayson J.A. The appellant complains that Finlayson J.A. applied a standard of “simply too high” (p. 661). It is true that he thought the award was too high, but that observation must be understood in light of other comments made in the course of his reasons. Finlayson J.A. concluded there was “no justification for such a radical departure from precedent” (pp. 661-62), which revealed awards in the range of $7,500 to $15,000. In his view, an appropriate figure for quantum requires a “balancing of factors such as those enumerated by Blackmun J.” (p. 667) in Pacific Mutual Life Insurance, supra. Finlayson J.A. looked at “the degree of reprehensibility of the defendant’s conduct” (p. 666) and concluded that “[t]his case does not demonstrate that there was such insidious, pernicious and persistent malice as would justify an award of this magnitude” (p. 666).

135 With respect to precedent, it must be remembered that the respondent’s trial counsel objected to any range or “bracket” of appropriate figures being given to the jury. Had the jury been given the information, it may have influenced their views. The respondent itself appears to have been unimpressed by the size of prior awards of punitive damages. In its factum, commenting on Crabbe’s letter of June 9, 1994, counsel states, “However, it should also be noted that Mr. Crabbe was clearly attempting to allay Pilot’s concern about the Whiten’s bad faith claim at a time when punitive damage awards against insurers were in the range of $7,500.00 - $15,000.00.” Pilot’s concern may have been easy to allay when the expected exposure to punitive damages was only $15,000.

136 The respondent objects that, prior to this judgment, the highest previous award in an insurer bad faith case was $50,000. However, prior to the $800,000 award of punitive damages upheld in Hill, supra, the highest award in punitive damages in a libel case in Canada was $50,000: Westbank Band of Indians v. Tomat, [1989] B.C.J. No. 1638 (QL) (S.C.). One of the strengths of the jury system is that it keeps the law in touch with evolving realities, including financial realities.

137 Finlayson J.A.’s central point was the need to pay close attention to the “balancing of factors” (p. 667), and in particular, the “reprehensibility of the defendant’s conduct” (p. 666). There was evidence to support the view expressed by Finlayson J.A., but there was also evidence to support the contrary view of Laskin J.A., in dissent (at p. 659):

. . . Pilot acted maliciously and vindictively by maintaining a serious accusation of arson for two years in the face of the opinions of an adjuster and several experts it had retained that the fire was accidental. It abused the obvious power imbalance in its relationship with its insured by refusing to pay a claim that it knew or surely should have known was valid, and even by cutting off rental payments on the Whiten’s rented cottage. It took advantage of its dominant financial position to try to force the Whitens to compromise or even abandon their claim. Indeed, throughout the nearly two years that the claim was outstanding, Pilot entirely disregarded the Whitens’ rights.

138 It seems to me, with respect, that this disagreement among very senior appellate judges turns on precisely the factual issues and inferences that were remitted to the jury for its determination. I would hesitate to characterize the considered opinion of any experienced and learned appellate judge as not only wrong but “irrational”.

139 Moreover, the trial judge, who sat through the evidence, went out of his way to comment that the award, while high, was reasonable. It was rational. He then added to Pilot’s burden an award of over $320,000 in solicitor‑client costs.

140 Having accepted with some hesitation the adequacy of the trial judge’s instructions to the jury, and there being no convincing demonstration that the jury’s subsequent imposition and assessment of punitive damages were irrational, I would affirm the award of punitive damages.

IV. Conclusion

141 I would allow the appeal and restore the jury award of $1 million in punitive damages, with costs in this Court on a party‑and‑party basis.

142 The respondent’s cross-appeal against the award of any punitive damages is dismissed with costs to the appellant, also on a party‑and‑party basis.

The following are the reasons delivered by

143 LeBel J. (dissenting on the appeal) – This case raises important issues about the proper functions of tort law, the role of punitive damages within it and the control of jury awards. In the end, I respectfully disagree with Binnie J.’s reasons and proposed disposition of the appeal. Although I agree that the bad faith of the Pilot Insurance Company (“Pilot” or “Pilot Insurance”) in its handling of the claim, up to and during the trial, amply justifies awarding punitive damages, an award of $1 million goes well beyond a rational and appropriate use of this kind of remedy, especially in what began as a problem of contract law. The majority of the Court of Appeal for Ontario set the amount of punitive damages at a sum which was consistent with the nature and purpose of punitive damages in the law of torts and had sufficient reasons to interfere with the jury’s verdict. I find myself in substantial agreement in this respect with Finlayson J.A.’s reasons in the Court of Appeal.

144 In these reasons, I will not attempt a full review of the facts that gave rise to this litigation. My colleague has made a careful review of them and I will refer to them only inasmuch as may be required by some particular aspect of this appeal.

1. Punitive Damages and the Functions of Tort Law

145 In this case, the plaintiff took action in order to claim compensation for a loss of about $345,000 under a residential and home insurance policy. The plaintiff did not ask for aggravated damages, but sought punitive damages. At the end of the trial, she was awarded the full amount of her actual loss, her costs on a solicitor‑client basis and punitive damages amounting to about three times her losses.

146 Tort law fulfills diverse functions. While deterrence and denunciation both still play a role, since it broke away from criminal law in the Middle Ages, in its core function, tort law has been compensatory or corrective. (See A. M. Linden, Canadian Tort Law (6th ed. 1997), at pp. 4-7.)

147 The purpose of this part of our legal system remains to make good the loss suffered, no less, no more:

The general principles underlying our system of damages suggest that a plaintiff should receive full and fair compensation, calculated to place him or her in the same position as he or she would have been had the tort not been committed, in so far as this can be achieved by a monetary award.

(Ratych v. Bloomer, 1990 97 (SCC), [1990] 1 S.C.R. 940, at p. 981, per McLachlin J. (as she then was); see also Jacobi v. Griffiths, 1999 693 (SCC), [1999] 2 S.C.R. 570, at para. 70, per Binnie J.)

148 The award of punitive damages in discussion here leads us far away from this principle. It tends to turn tort law upside down. It transmogrifies what should have remained an incident of a contracts case into the central issue of the dispute. The main purpose of the action becomes the search for punishment, not compensation. Perhaps, at some time in the future, this will be viewed as part of a broad intellectual and social movement of privatization of criminal justice, consonant with the general evolution of society. For the time being, without using in terrorem arguments, such an award has a potential to alter significantly what would appear to have been the proper function of tort law.

149 It would be useless, within the limits of these reasons, to review the controversies surrounding punitive damages. Whatever the doubts raised at times by academic or judicial opinion about their place in the law of torts, Canadian tort law now accepts that a private action in torts may give rise, in the proper cases, to an award of punitive damages. They may even be awarded in a contracts case, provided an independent actionable wrong is made out. (See Vorvis v. Insurance Corp. of British Columbia, 1989 93 (SCC), [1989] 1 S.C.R. 1085, at p. 1104; Wallace v. United Grain Growers Ltd., 1997 332 (SCC), [1997] 3 S.C.R. 701, at para. 79, per Iacobucci J.)

150 The use of this remedy appears to be flexible. Nevertheless, if necessary, punitive damages must be granted and assessed in such a way that they serve rational purposes and represent a proportionate response to the behaviour of defendants and the harm caused. Also, as can be seen in this case, when the matter comes up before a civil jury, jurors must be instructed properly as to the nature and purposes of punitive damages and the conditions under which they may be awarded.

2. Rationality and Proportionality

151 I agree with Binnie J. on the core principles governing the award of punitive damages. The key considerations remain the rationality and proportionality of the award. The concept of rationality remains grounded in the nature of tort law, its historical development and the functions it is now playing in modern society. First and foremost, the assessment of damages should not lead, in some cases, to a confusion of criminal law and private law principles, given that punitive damages and criminal punishment target primarily the conduct of the respondent or accused and are not primarily concerned with making good the loss or harm suffered by victims. The main concern of punitive damages remains the preservation of public order, and the assuaging of such harm as may have been done to the public good and to the social peace.

152 At a time when the King’s writ did not always run throughout the realm, losses or harm to a person or property were made good by punishment or alternative measures, which were usually payment or compensation. For example, in early Anglo-Saxon England, the law knew very few strictly capital crimes. In most cases, two sets of punishment were available. On the one hand, the law allowed for a system of revenge. The victim, when possible, or the kin, would exact retribution on something like an “eye for an eye” principle. On the other hand, it was often seen as undesirable to permit blood feuds to go on endlessly. Thus there emerged a system of compensation by which crimes could be atoned for, by paying a victim or the victim’s kin some sum, dependent upon the type of act committed and the class or status of the victim. These sums were even codified and often known as wergeld or blood money. (See C. Hibbert, The Roots of Evil: A Social History of Crime and Punishment (1963), at pp. 3-5; W. J. V. Windeyer, Lectures on Legal History (2nd ed. rev. 1957), at p. 17.) During the Middle Ages, a clear distinction was drawn between pleas of the Crown which gave rise to a criminal law action and common pleas which were actions for damages. The merger of criminal law and the law of civil responsibility thus came to an end. (See Windeyer, supra, at p. 63.) One branch of the law aimed at punishment and preservation of the public; the other sought to make good losses suffered by the victim of some specific tort. Since then, in the common law, tort law has been viewed primarily as a mechanism of compensation. Its underlying organizing structure remains grounded in the principle of corrective justice, although policy concerns may play at times a considerable part in determining the outcome of a particular case, as, for example, in actions based on the tort of negligence. (See, for example, Edwards v. Law Society of Upper Canada, [2001] 3 S.C.R. 562, 2001 SCC 80; Cooper v. Hobart, [2001] 3 S.C.R. 537, 2001 SCC 79; Kamloops (City of) v. Nielsen, 1984 21 (SCC), [1984] 2 S.C.R. 2.)

153 A private tort action usually focusses on undoing the wrong perpetrated by a specific defendant against a particular plaintiff. For this reason, in the law of tort, “[p]roof of negligence in the air, so to speak, will not do” (Pollock, Torts, 11th ed., p. 455, quoted in Palsgraf v. Long Island R. Co., 162 N.E. 99 (N.Y. 1928), per Cardozo C.J.). The liability of the defendant and the right of recovery of the plaintiff do not exist independently. In this respect, an obligation arising out of torts remains as relational as in the law of contracts: “not only must the contractually bound defendant perform the promised act, but that performance is owed to a particular plaintiff” (E. Weinrib, The Idea of Private Law (1995), at p. 52).

154 Given the relational nature of the wrong committed by this defendant against the plaintiff, the remedies chosen by the court must remain consistent with this basic characteristic. The defendant must pay damages to the plaintiff in order to undo, inasmuch as can be done, the wrong caused. This principle governs the assessment of the quantum of damages as a rule. Within the limits set by principles of legal policy or by such conditions as specific torts may attach to their recovery, damages should correspond to the amount required to put the plaintiff in the position he or she would have been in, but for the wrong committed by the defendant. This principle of corrective justice will be implemented more easily in situations where a loss is fairly easy to ascertain. In other fields of the law, compensation remains at best an approximation, grounded, as it must often be, on a reasonable guess as to the nature or extent of the impact of a particular wrong. In the case of personal injury, the problem is well known but has defied just about every effort to find clear and logical solutions. Even in such cases, the concern remains the same. If the ultimate goal of perfect justice remains beyond the reach of the courts, tort law seeks to compensate as fully as possible an actual loss caused by a specific defendant to a particular plaintiff.

155 In the development of the principles governing the compensation of economic loss, the analytical methods adopted by the Court since Canadian National Railway Co. v. Norsk Pacific Steamship Co., 1992 105 (SCC), [1992] 1 S.C.R. 1021, reflect the relational nature of torts at common law. The concern to control recovery through the concepts of proximity and foreseeability signals the importance of this key principle that tort law is not grounded in responsibility at large to society or to indeterminate classes of plaintiffs, but on obligations to compensate adequately victims standing in a close enough relationship with a tortfeasor.

156 By reason of the relational nature of private tort law, punitive damages do not fit easily into its overall scheme, such as it may be. They are considered at a point in a suit when, in principle, the complainant has suffered a wrong but has been granted compensation as complete as the law allows in order to put him or her back in his or her former position. Hence, punitive damages come on top of everything else and carry no particular price tag. At the same time, an award of punitive damages may reflect broader and different societal concerns. These concerns reflect their position in the law: they are designed to punish, not to compensate. An award of compensatory damages may, in a way, punish the defendant due to the very fact that he or she has been found in breach of some legal duty, that he or she is ordered to indemnify a plaintiff and that he or she has had to go through the inconvenience of a trial and also sometimes the humiliation of adverse publicity attached to legal action. Indeed, by itself, an award of general damages may be punishment enough. It does not mean, though, that an action is primarily punishment: the compensatory nature of the claim remains. Punitive damages differ strikingly from all other damages as the sole reason for awarding them is to punish, as Professor Feldthusen has pointed out. Even aggravated damages differ in this respect from punitive damages (see B. Feldthusen, “Recent Developments in the Canadian Law of Punitive Damages” (1990), 16 Can. Bus. L.J. 241).

157 Aggravated damages served the traditional corrective purpose of the common law: to make the plaintiff whole for injuries to interests that are not properly compensable by ordinary damages. Punitive damages target not loss, but conduct. (See Vorvis, supra, at pp. 1098-99; Hill v. Church of Scientology of Toronto, 1995 59 (SCC), [1995] 2 S.C.R. 1130, at para. 196.) The defendant’s wrong must then be considered directly and separately in order to assess its severity and, accordingly, the appropriate degree of punishment. The other forms of damages look to the loss of the plaintiff, but punitive damages refer essentially to the degree of culpability of the defendant’s action.

158 The difficulties inherent in the nature of punitive damages have given rise to doubts as to their proper place in the law of torts. Some critics have indeed opined that they have no place in the proper structure of tort law, equating it with some form of “palm tree justice”. (See Cassell & Co. v. Broome, [1972] A.C. 1027 (H.L.), at p. 1087, per Lord Reid.) Under the pine trees of this country, as we know, punitive damages have found a place in the law of torts. Nevertheless, as with any legal institution, punitive damages must address some identifiable purpose and concern in order to define their proper role. An overriding objective of general deterrence remains problematic, if punitive damages are to remain a useful incident of tort law. Otherwise, their use may turn some parts of the law of tort into a sort of private criminal law, devoid of all the procedural and evidentiary constraints which have come to be associated with the criminal justice system.

3. The Context of the Award of Punitive Damages

159 At this point, I must turn from principle to some particular facts of this case. I will try to identify which broad social purpose could be addressed by a large punitive damages award. This case started as litigation based on a home insurance contract. The insured had the misfortune to run across what might be characterized as the insurer from hell. Far from finding peace of mind and receiving a settlement in a fair and diligent manner, the claimant faced obduracy and bad faith. No evidence has been offered, though, that such conduct was a regular incident of Pilot Insurance’s way of running its business. It looks more like a rogue file mishandled by an overeager manager, aided and abetted by counsel who seemed to have misunderstood the nature of his duties as an officer of the court and as a member in good standing of the Law Society of Upper Canada. Indeed, what is referred to at para. 16, in the reasons of my colleague, as a concession in respondent’s factum that the upper management of the insurer was fully apprised of what was going on in the Whiten file, describes a line of reporting authority. Neither the so‑called concession nor the evidence referred to in para. 17 of Pilot’s factum establishes that the upper management of the company had acquired actual knowledge of the mishandling of the Whiten claim by its local manager and the lawyer he had retained. In addition, no suggestion has been made that such behaviour is widespread in the insurance industry in Ontario or elsewhere in Canada.

160 What, then, are the purposes of an award of punitive damages in this context? Once the purposes are determined, what should be a reasonable and proportionate award? Some might think that the prospect of punitive damages could perhaps strike fear into the hearts of cold-blooded bean counters lurking somewhere in the basement of insurance companies’ head offices. The terrible swift sword of the law might draw blood from a company used to turning a blind eye to abuses by its middle managers. In the absence of evidence about the flaws of the entrepreneurial culture of Pilot Insurance, the situations or the particular evils rampant in the insurance industry, if any, what is left is a desire to punish adequately acts of bad faith and unfair dealing by a manager and counsel of an insurance company. (As my colleague indicates, this remark does not apply to counsel who represented Pilot in the Court of Appeal and in this Court.)

161 In this case, the sole narrow purpose of an award for punitive damages appears to be the punishment of the bad faith of the insurer in the discharge of its duties under what should be a good faith contract on both sides. The insurer must compensate in a timely manner. It has the right, even the duty, to investigate claims, but must do so fairly and diligently. For his or her part, the insured must file his or her claims promptly and assess his or her losses as accurately as he or she can. Given the nature of the contract, bad faith may constitute an actionable wrong and attract the sting of punitive damages. The challenge remains to assess them properly, in a manner consistent with the basic purposes of tort law. In the present appeal, an award of damages must reflect first of all the narrow objective defined by the facts of this case; as well, it must also reflect the relational nature of tort law, although, at times, such an award may be viewed as a deterrent to others. (See Hill, supra, at para. 196, per Cory J.; Vorvis, supra, at p. 1108, per McIntyre J.) But the need for general deterrence is far from clear in the present case. The requirement of a proper connection between award and conduct requires a close fit between the amount of the award and the misbehaviour of the respondent. An important consideration remains the nature of the dispute, which arose in the context of a contractual relationship concerning well-defined economic interests and not with respect to moral or dignity interest as in the case of an action for defamation. In addition, concerns about industry practices should mainly be addressed through the appropriate regulatory and penal regimes, rather than through haphazard punitive damages awards. (See Ontario Law Reform Commission, Report on Exemplary Damages (1991), at p. 37.)

162 In the case of disputes concerning damage to property or economic interests, the retributive aspect of the law should not play a major role in litigation. Granting an indemnity of about three times the compensation for loss of property under an insurance policy fulfills no rational function. Despite the moral satisfaction we may derive from giving a good whack to an insurance company and some misguided middle managers, the verdict of the jury does not much advance the case of sound and fair management in the insurance industry. The award fails the rationality test because its sole purpose remains to punish adequately bad faith and unfair dealing by employees of Pilot and its counsel. It does not address any widespread practice in the insurance industry. It does not pretend to effect a disgorgement of unfairly acquired profit. The punishment far exceeds whatever property or economic losses may have been caused by the nonperformance of the contract. In such cases, the criterion of proportionality requires that the use of punitive damages remain carefully controlled and that punitive damages should not significantly exceed the amount of damages to property, or economic interests, including aggravated damages, if any are claimed.

4. Control of Awards

163 While Canadian courts have not imposed formal caps on punitive damages, the discretionary nature of such awards and the difficulties of setting amounts in a rational manner indicate that proper instructions should be given to juries about factors and methods to be used in the difficult task of assessing punitive damages. In my respectful view, that is exactly the concern that the Ontario Court of Appeal attempted to address. Finlayson J.A. not only found that the award was too high, but that it was so high as to be unreasonable and to require interference by the Court of Appeal. The majority adopted a figure which appears reasonable and proportionate. It imposed significant punishment for the bad faith of Pilot without upsetting the proper balance between the compensatory and punitive functions of tort law.

164 Moreover, flexibility and discretion are not the only values at stake in the development of legal rules in the law of damages. Some degree of predictability and consistency should also be factored into situations where the nature of the damages suffered makes it difficult for a jury to determine a proper quantum, as in the case of non-monetary losses and personal injury litigation. In this area, it should be remembered that our Court deemed it necessary to impose formal caps or a “bright line”. In the 1978 trilogy Andrews v. Grand & Toy Alberta Ltd., 1978 1 (SCC), [1978] 2 S.C.R. 229; Thornton v. Board of School Trustees of School District No. 57 (Prince George), 1978 12 (SCC), [1978] 2 S.C.R. 267; and Arnold v. Teno, 1978 2 (SCC), [1978] 2 S.C.R. 287, at p. 334, the Court set a cap of $100,000 for damages which were intended as restitution for non-monetary losses after adjustment for inflation; our Court has kept this cap in place ever since. Indeed, in 1995, Sopinka J. wrote that the “trilogy has imposed as a rule of law a legal limit to non-pecuniary damages”. (See ter Neuzen v. Korn, 1995 72 (SCC), [1995] 3 S.C.R. 674, at para. 114.) In Andrews, supra, at p. 261, Dickson J.) (as he then was) found that some kind of fair and reasonable limit should be imposed given the difficulty of putting a price on non-pecuniary losses:

[T]he problem here is qualitatively different from that of pecuniary losses. There is no medium of exchange for happiness. There is no market for expectation of life. The monetary evaluation of non-pecuniary losses is a philosophical and policy exercise more than a legal or logical one. The award must be fair and reasonable, fairness being gauged by earlier decisions; but the award must also of necessity be arbitrary or conventional. No money can provide true restitution.

(See also Arnold, supra, at p. 331, per Spence J.)

165 In the end, imposing limits on such claims was not viewed as working an injustice on claimants. Setting punitive damages at amounts which do not exceed significantly the real economic loss, where the loss suffered concerns mainly property and economic interests, would leave them in their proper place within the scheme of the law of torts. At the same time, this approach would avoid undermining the structure of the law of torts and its core function.

5. Function of the Jury

166 In the circumstances of this case, the fact that the award was made by a jury does not make it more immune to appellate review; as appears from the record, the jurors were themselves uncomfortable with the problem of determining the proper quantum of punitive damages. They sought further instructions, but the trial judge left them with the “skeletal instructions” given during the charge. Hence, they ended up with an award which my colleague would not have given, and which he says stands at the outer limits of the reasonable, but within it. The Court of Appeal did only its duty when it reviewed this award. A verdict which breaches the requirements of proportionality and rationality required appellate review.

167 Some problems may arise also from aspects of the practice followed at trial. According to the present Ontario practice, if one counsel objects, the presiding judge will not instruct the jury as to the range of awards save in the case of personal injuries. (Courts of Justice Act, R.S.O. 1990, c. C.43, s. 118; Holmested and Watson: Ontario Civil Procedure (loose-leaf ed.), vol. 1, at p. CJA-242; Caron v. Chodan Estate (1992), 58 O.A.C. 173, at p. 175, per Arbour J.A. (as she then was); P. G. Vogel, Cohen Melnitzer’s Civil Procedure in Practice (loose-leaf ed.), vol. 1, at p. 12-20; Gray v. Alanco Developments Ltd., 1967 221 (ON CA), [1967] 1 O.R. 597 (C.A.); Howes v. Crosby (1984), 1984 2172 (ON CA), 45 O.R. (2d) 449 (C.A.).)

168 The problems that occurred in the present case demonstrate that some sort of instruction on the range of punitive damages awards, even without counsel’s agreement, would have been useful. Without removing the jury’s discretion, it would at least communicate to them some idea of past figures and of guidelines that may be found in appellate or Supreme Court of Canada judgments. They should also be instructed clearly that an award of general damages may also amount to all the punishment that is necessary in a given case. Failing this, and in the absence of a proper application of the rationality and proportionality criteria, problematic awards are bound to happen. Courts should take care that they do not alter the nature of tort law by turning the focus of civil litigation away from compensation of a claim to punishment of defendants.

169 For these reasons, I would dismiss the appeal without costs and the cross-appeal with costs.

Appeal allowed with costs, LeBel J. dissenting. Cross-appeal dismissed with costs.

Solicitors for the appellant/respondent on cross-appeal: MacMillan Rooke Boeckle, Toronto.

Solicitors for the respondent/appellant on cross-appeal: Lerner & Associates, Toronto.

Solicitors for the intervener the Insurance Council of Canada: Davies, Phillips & Vineberg, Toronto.

Solicitors for the intervener the Ontario Trial Lawyers Association: Ross & McBride, Hamilton.