AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2020 >> [2020] MelbULawRw 8

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Maher, Felicity --- "An Empirical Study of Exemplary Damages in Austra" [2020] MelbULawRw 8; (2020) 43(2) Melbourne University Law Review 694


AN EMPIRICAL STUDY OF EXEMPLARY DAMAGES IN AUSTRALIA

Felicity Maher[1]*

In a series of statutory interventions concentrated around the early 2000s, Australian parliaments legislated to limit or exclude the availability of exemplary damages. This prompted the authors of a recent empirical study of exemplary damages in the United Kingdom to claim that the award of exemplary damages in Australia has been ‘rendered effectively extinct’. Inspired by their study, this article is an empirical study of exemplary damages in Australia. The study examines cases decided between 2000 and 2016 in all Australian jurisdictions, at all levels of court the decisions of which are available electronically. Using appropriate methods of statistical analysis, the study considers the number of claims for exemplary damages, the frequency of awards, and the size of those awards. Various features of the claims in which the awards were made are also considered, including the nature of the cause of action and the defendant, and the other remedies awarded. The main conclusion of the study is that exemplary damages are alive and well in Australia.

Contents

I Introduction

Exemplary damages are a controversial — perhaps the most controversial — private law remedy. Exceptionally, their primary functions are punishment and deterrence.[1] Various objections are commonly made against exemplary damages. It is said that punishment is and should be the exclusive province of the criminal law, that civil law does not provide adequate safeguards to defendants in the imposition of punishment, that exemplary damages provide a windfall to the plaintiff, and that they encourage litigiousness.[2] Courts and commentators frequently express concern about the overuse of exemplary damages and excessive awards. They repeatedly emphasise that exemplary damages are rare and should be awarded only exceptionally,[3] and that in making awards of exemplary damages, courts should exercise restraint and moderation.[4] Moreover, in Australia, both federal and state and territory parliaments have legislated to limit or exclude the availability of exemplary damages in certain circumstances.[5]

This statutory intervention prompted the authors of a recent empirical study of exemplary damages in the United Kingdom (‘UK Study’) to suggest that ‘since approximately 2001, the award of punitive damages has been rendered effectively extinct in Australia’.[6] The UK Study was the first of its kind in that jurisdiction, or elsewhere in the common law world.[7] In Australia, there has been no empirical study of exemplary damages, nor even a report by a law reform body.[8] This is despite repeated calls for such a study.[9]

Inspired by the UK Study, and especially its suggestion about the Australian position, this article comprises an empirical study of exemplary damages in Australia. It largely adopts the methodology of the UK Study but, as will be discussed later,[10] also differs in some significant respects. Like the UK Study, the main questions considered by this study concern the number of claims for exemplary damages in Australia, the rate of success in those claims, the amount of exemplary damages awards and how these vary across different jurisdictions within Australia, and different categories of cause of action and defendant. The study also examines whether there is any relationship between exemplary damages and aggravated and compensatory damages. More generally, the study asks whether the frequently expressed concerns about exemplary damages are warranted, and whether there is a sound basis for either the existing statutory intervention in this field, or any future legislative reform.

As will be seen, the main conclusion of the study is that exemplary damages are alive and well in Australia. Indeed, they are in more robust health here than in the UK. In particular, over the period of the study, exemplary damages were sought in 253 claims and awarded at an overall success rate of just less than 50%. The mean award was just over $105,000. By far the greatest number of claims were brought in New South Wales. Exemplary damages were most often awarded where the claim included both interference with the person and interference with property. The greatest awards were also made in interference with property claims. More successful claims were made against public bodies than any other category of defendant; of these claims, more than three quarters were made against police. Finally, exemplary damages tend to go hand in hand with aggravated damages and, when assessing exemplary damages, courts take into account the amount of any compensatory damages.

Part II of this article sets out a brief summary of the legal principles governing exemplary damages in Australia. Part III summarises the key legislative provisions limiting or excluding the availability of exemplary damages in this country. Part IV describes the methodology adopted for the empirical study and how it differs from that of the UK Study. Part V sets out the results of the study. Finally, Part VI analyses the results, considering what they tell us about exemplary damages in Australia standing alone, and in comparison with the UK.

II Legal Principles Governing Exemplary Damages in Australia

The long history of exemplary damages in English common law has been described in detail by various commentators.[11] In Australia, common law exemplary damages were first considered by the High Court in 1920. Whitfeld v De Lauret & Co Ltd (‘Whitfeld’)[12] recognised the availability of the remedy and laid down a test for its award: where the defendant has undertaken ‘conscious wrongdoing in contumelious disregard of another’s rights’.[13] The next major development was in 1966, when the High Court in Uren v John Fairfax & Sons Pty Ltd (‘Uren’)[14] decided that English law’s restrictive approach to exemplary damages, originating from the decision in Rookes v Barnard (‘Rookes’),[15] did not apply in Australia.[16] In Rookes, the House of Lords was invited to abolish exemplary damages in England. Given the long history of the remedy, their Lordships declined to do so, instead confining the scope of exemplary damages to the minimum possible within the limits of authority. Most previous cases of ‘exemplary damages’ were reclassified as really awarding aggravated damages. As for the rest, the House of Lords identified only three categories of case in which exemplary damages were available — one statutory and two common law. The two common law categories were: (i) where there were ‘oppressive, arbitrary or unconstitutional acts by servants of the government’; and (ii) where the defendant cynically calculated that the profit to be made by their conduct might well ‘exceed the compensation payable to the plaintiff’.[17] In Rookes, the House of Lords also emphasised that, in assessing exemplary damages, the watchword was moderation.[18]

In Australia, a claim for exemplary damages need not fall within a Rookes category. Provided the test first stated in Whitfeld is satisfied, exemplary damages may be available in answer to any cause of action.[19] The so-called ‘cause of action test’ — also applied (for a time) in England[20] — was never adopted in Australia. That test required that for exemplary damages to be available, the cause of action relied upon must be one for which exemplary damages were available before Rookes.[21]

Since Uren, claims for exemplary damages have been before the High Court on several occasions. It has been suggested that the modern tendency in Australia is to ‘embrace, or at least to concede’ exemplary damages.[22] The Whitfeld test continues to be applied.[23] The position today is that, subject to statutory intervention, exemplary damages are available in principle for any type of claim, including unintentional torts such as negligence.[24] Apart from statute, the only real exceptions are claims for breach of contract[25] and equitable wrongdoing.[26] Exemplary damages may be available even where the defendant is not the wrongdoer, but (for example) the wrongdoer’s insurer,[27] or vicariously liable employer.[28] This is on the basis that, although the punishment and deterrence functions of exemplary damages might not be achieved in such a case, other functions of the remedy will be, including preventing plaintiffs from taking revenge.[29] Exemplary damages will not be available, however, where substantial punishment has been imposed on the defendant by the criminal law.[30] In the assessment of exemplary damages, all relevant aggravating and mitigating factors are taken into account,[31] and the court will also consider the means of the defendant.[32] There is ‘no necessary proportionality’ between exemplary and compensatory damages.[33] But like English courts, Australian courts must exercise moderation in assessing exemplary damages.[34]

III Statutory Intervention

There has been significant statutory intervention in Australia limiting or excluding the availability of exemplary damages in certain circumstances. The earliest legislation dates from the 1930s. But the activity was concentrated around the late 1980s and the 2000s. The table annexed to this article summarises the key legislative provisions.[35]

The earliest provisions concern the survival of claims. Adopting 1934 legislation in the UK,[36] enactments from 1935 in Tasmania[37] and other jurisdictions in the years that followed[38] provided that where a cause of action survives for the benefit of the estate of a deceased person, the damages recoverable for the benefit of the estate do not include exemplary damages.

Thereafter, and especially during the late 1980s and the 2000s, statutory intervention occurred in three main areas: defamation, personal injury (generally, and arising from motor vehicle accidents and in the workplace), and the liability of the state for the tortious conduct of police. There was, however, no legislation considering the availability of exemplary damages generally. This may be a function of the fact that no Australian law reform body — federal, state or territory — has considered this issue.[39]

In relation to defamation, New South Wales (home to Sydney, which has ‘a good claim to ... being the defamation capital of the world’)[40] was the first to legislate. Section 46(3) of the Defamation Act 1974 (NSW) abolished exemplary damages in defamation actions. It provided: ‘In particular, damages for defamation: (a) shall not include exemplary damages’. Thirty-odd years later, with the enactment of the uniform defamation Acts, every other state and territory followed.[41]

In relation to personal injury generally, the Review of the Law of Negligence (‘Ipp Report’) recommended in 2002 that exemplary damages should be abolished in claims for negligence causing injury or death.[42] This was on the basis that exemplary damages confused the functions of the civil and criminal law, amounted to an undeserved windfall to the plaintiff, and were often unpredictable and excessive in amount.[43] Further: ‘The patchwork of legislation now in force limiting or abolishing exemplary damages in various types of case can be taken to reflect a community view that the remedy of exemplary damages is neither necessary nor desirable.’[44] Following the Ipp Report, civil liability Acts were enacted in various jurisdictions, excluding exemplary damages in claims for negligence resulting in death and personal injury.[45] For example, s 21 of the Civil Liability Act 2002 (NSW) provides: ‘In an action for the award of personal injury damages where the act or omission that caused the injury or death was negligence, a court cannot award exemplary or punitive damages or damages in the nature of aggravated damages.’ To the same effect, and in similar terms, was s 87ZB of the Trade Practices Act 1974 (Cth), enacted in 2004.[46]

As highlighted in the Ipp Report, these reforms were predated[47] by legislative provisions in most states and territories excluding exemplary damages in claims for personal injury arising from a motor vehicle accident.[48] For example, s 81A of the Motor Accidents Act 1988 (NSW) provided: ‘A court shall not award exemplary or punitive damages to a person in respect of a motor accident.’ Several such provisions take the form of excluding the liability of a compulsory third-party insurer for exemplary damages.[49] Similarly, the civil liability reforms were also predated[50] by legislative provisions in several jurisdictions excluding exemplary damages in claims for workers’ compensation.[51] For example, s 151R of the Workers Compensation Act 1987 (NSW) provided: ‘A court may not award exemplary or punitive damages to a person in an award of damages.’ Again, some of these provisions take the form of excluding liability of an insurer for exemplary damages.[52]

It has been said that many claims for exemplary damages today are brought against the police,[53] and, further, that exemplary damages are most frequently awarded in such actions.[54] As will be seen,[55] this claim is borne out by the results of this study. This is notwithstanding statutory intervention in the Commonwealth and several states and territories excluding the vicarious liability of the state for exemplary damages in claims arising out of the tortious conduct of police.[56] For example, s 137(6) of the Police Act 1892 (WA) provides: ‘The Crown’s liability under subsection (5) does not extend to exemplary or punitive damages.’

IV Methodology

The methodology adopted in this study largely replicates that of the UK Study, with some significant differences that are identified in this part.

A Scope

In relation to temporal scope, cases decided in the time period 2000 to 2016 inclusive were included. This reflects the UK Study, which included cases decided in the period 2000 to 2015 inclusive. The reasons given by the authors of the UK Study for choosing that period of time are that it enabled an examination of the ‘contemporary operation of the law’ of exemplary damages, the 16-year period produced a large enough sample to permit the drawing of ‘meaningful conclusions’, and, as from about 2000, there was a ‘dramatic increase’ in the availability of cases electronically.[57] These reasons apply with equal force here. Further, adopting the same temporal scope in this study (with the addition of one extra year) will allow for some useful comparative analysis of the Australian and UK positions. The period 2000 to 2016 also encompasses or postdates the most concentrated period of statutory intervention in this country. In addition, the decision to limit the scope of the study in this way was a pragmatic decision, to keep the number of cases to a manageable size.

In relation to institutional and jurisdictional scope, the study includes cases decided by every court (but not tribunal)[58] of the Commonwealth and every state and territory, applying Australian law, where those cases were available electronically.[59] Unlike the UK Study, which examined only first instance decisions, this study also considers appeals.[60] So where a first instance decision on exemplary damages was appealed, the case on appeal was included, rather than the first instance decision (or any intermediate appellate decision). In this way, the sample represents the final results in all cases included. It is considered that this benefit outweighs the concern that appealed cases may sometimes be atypical.

B Case Searches

Within these parameters, case searches were conducted using Lexis Advance, by entering the phrases ‘exemplary damages’ and, separately, ‘punitive damages’ into the full text search box.[61] Results were crosschecked using Westlaw AU and additional cases included. These searches yielded in excess of 1,880 cases.[62] All cases were reviewed[63] to identify those which satisfied the following four criteria: (i) liability at common law was established or accepted; (ii) common law exemplary damages were sought by the plaintiff; (iii) common law exemplary damages were awarded or refused on factual grounds; and (iv) the case finally disposed of the claim for exemplary damages.[64] These criteria may be slightly more refined than those of the UK Study. They ensure that this is a study of common law exemplary damages in Australia, rather than equivalent or similar statutory remedies.[65] They ensure that the study captures only cases in which common law exemplary damages were available in principle, but awarded or refused on factual rather than purely legal grounds. The latter mainly arise under statute where, as discussed above,[66] legislative provisions exclude the availability of exemplary damages in certain circumstances.[67] But they also arise at common law: for example, at common law, exemplary damages are not available for breach of contract.[68] As in the UK Study, these criteria also exclude cases where the court found against the plaintiff on liability, but proceeded to consider exemplary damages in case it was wrong.[69]

This review reduced the number of cases included to 186.[70] All were accessed directly, rather than through an appellate decision, given that appeals were included.[71]

C Coding the Cases

All 186 cases were reviewed personally by the author. Data from the cases was then entered into a series of spreadsheets. Separate spreadsheets were produced for each jurisdiction and for each court in that jurisdiction. Following the UK Study, separate entries were made for each claim within a case. So, for example, if a case involved two plaintiffs, each claiming exemplary damages against a single defendant, two entries were made — one for each plaintiff’s claim. Or, where a single plaintiff sought exemplary damages against two defendants, two entries were made — one for the plaintiff’s claim against each defendant. This increased the sample from 186 cases to 253 claims.

There were two exceptions to this approach.[72] The first was where multiple parties were in substance a single party (for example, an employee and a vicariously liable employer, or an insured and an insurer). The second was where multiple defendants were held jointly and severally liable for the same wrong, such that a single exemplary damages award, to be paid only once, was made against all defendants. The advantages of this approach, as noted by the authors of the UK Study, are that it enables more accurate description of what happens in cases involving more than one plaintiff or defendant, and it increases the sample size.[73] However, as the same authors also note, the approach adds ‘an issue of significant complexity’.[74] Under UK law, where two or more plaintiffs are entitled to exemplary damages against a single defendant, authority makes clear that courts should make one award of exemplary damages and divide it between the plaintiffs.[75] Using the counting claims (rather than cases) approach, however, although capturing the amount received by each plaintiff, would not capture the total amount paid by the defendant. The UK Study dealt with this issue by recording in the spreadsheet both the amount of exemplary damages each plaintiff was entitled to receive, and the amount the defendant was required to pay. Overall, however, the authors noted that doing this had little impact on the statistical analysis: ‘the different ways of measuring the quantum of ... [exemplary damages] awards was unimportant in the context of this study’.[76]

A somewhat different approach was taken here. As mentioned, as in the UK Study, cases were broken down into claims, and separate entries made in the spreadsheets for each claim. However, how the case was broken down was a function of how it was dealt with by the court. For example, in a case with multiple plaintiffs but only one defendant, if the court made a separate award of exemplary damages for each plaintiff,[77] the case was broken down into the claims made by each plaintiff. Or, for example, in a case with multiple defendants and only one plaintiff, if the court made a separate award of exemplary damages against each defendant,[78] the case was broken down into the claims made against each defendant. Finally, in a case where multiple causes of action were relied upon by the plaintiff,[79] if the court made a separate award of exemplary damages for each cause of action, the case was broken down into separate claims in respect of each. Accordingly, for each claim, what was recorded was the amount (if any) of exemplary damages awarded for that claim. Unlike the UK Study, this study did not record separately the amount received by the plaintiff and the amount paid by the defendant. This was considered appropriate, given the greater willingness of Australian courts than courts in the UK to make separate awards of exemplary damages to each of multiple plaintiffs, and against each of multiple defendants — as well as in respect of each of multiple causes of action.[80] So the ‘issue of significant complexity’[81] identified by the authors of the UK Study is of less significance in Australia. Further, given the results of the UK Study,[82] it is unlikely that not recording separately the amounts of exemplary damages received and paid would have made a significant difference to the statistical analysis.

The study then largely adopted the coding variables used in the UK Study. The following information was extracted from the cases and the columns of the spreadsheets populated accordingly:

Column A: claim number. Each claim was given a unique identifying number.

Column B: name of case.

Column C: citation.

Column D: date of decision. For each court, within each jurisdiction, claims were arranged chronologically.

Column E: jurisdiction (Commonwealth, state or territory).

Column F: court.

Column G: defendant category. Following the UK Study, three categories of defendant were used: (i) individual; (ii) public body; and (iii) corporation.

Column H: defendant subcategory. This column was principally used to identify claims brought against police. It was apparent from an early stage of the case review that a large number of claims for exemplary damages were brought against police, making this subcategory of defendant worthy of separate analysis. Further, all claims against police were treated as claims against the third category of defendant, public body. This was so even where exemplary damages were claimed against an individual police officer (instead of, or in addition to, the state on the basis of vicarious liability for the conduct of the officer). This was because, even in these cases, an award was made or refused against the individual officer for conduct in exercise, or purported exercise, of the powers of a police officer — that is, a public power. Further, separate analysis of claims made against police required analysis of all such claims, whether brought against the individual officer, the state, or both.

Column I: nature of the defendant’s wrong. This column set out a brief description of the defendant’s conduct.

Column J: cause of action. The particular cause or causes of action relied on by the plaintiff in the claim for exemplary damages were identified — for example, false imprisonment. Where there were more than one, all were separately identified.

Column K: cause of action category. Each cause of action relied upon by the plaintiff in the claim for exemplary damages was coded as falling within one of six categories of cause of action. For the purposes of comparative analysis, these categories were the same six used in the UK Study, namely: (i) interference with the person (for example, assault and false imprisonment); (ii) interference with property (for example, trespass to land); (iii) defamation and privacy invasions; (iv) abuse of power torts (for example, malicious prosecution); (v) economic torts (for example, deceit); and (vi) miscellaneous (for example, negligence). Again, where there was more than one cause of action, falling within more than one category of cause of action, all categories were separately identified. In this respect, the UK Study was not followed. There, ‘a judgment call was made as to the classification that seemed most appropriate’.[83] As will be seen,[84] this has important consequences for the statistical analysis. In this study, there were many claims that fell within more than one category of cause of action. There were a few options for dealing with this. First, in order to fit a claim within only one category of cause of action, as in the UK Study, a judgement call could have been made as to the most appropriate category. However, often this was simply not possible, as the claim fell unavoidably into two or more categories. For example, in claims against police, three causes of action were typically relied upon: assault, false imprisonment and malicious prosecution. The first two causes of action fall within the category of interference with the person but the third falls within the category of abuse of power torts. So this option was not considered satisfactory. Second, the claim could have been allocated arbitrarily to a single cause of action category. For example, where there were two applicable categories of cause of action, the first-mentioned could have been selected. However, this option was plainly unsatisfactory. Third, these claims might have been omitted from the study, but that would have distorted the results and was therefore also plainly unsatisfactory. Fourth, some further, combined categories of cause of action could be created — for example, a combined category of interference with the person and abuse of power torts. On the advice of a statistician, this approach was adopted.

Column L: whether exemplary damages were awarded (yes or no).

Column M: if so, the amount of exemplary damages awarded. All figures were adjusted for inflation to the year 2017 using the Reserve Bank of Australia’s online inflation calculator.[85]

Column N: whether aggravated damages were awarded (yes or no). Following the UK Study, this study also considers whether there is any relationship between exemplary and aggravated damages.

Column O: amount of aggravated damages awarded, adjusted for inflation.

Column P: amount of compensatory damages awarded, adjusted for inflation. Unlike the UK Study, what was reported in this column was the amount of compensatory damages excluding aggravated damages (if any).[86] Further, for the purposes of the statistical analysis, notwithstanding that aggravated damages are compensatory in function,[87] compensatory damages (excluding aggravated damages) were analysed separately from aggravated damages. This was on the basis that the circumstances for awarding aggravated and compensatory damages are different. The award of aggravated damages requires something more than loss on the plaintiff’s part: they are awarded when the harm done to the plaintiff by a wrongful act is ‘aggravated by the manner in which the act was done’.[88] However, there was one exception to this approach. As will be seen,[89] there were a number of cases in which the court did not separately specify the amounts of aggravated and compensatory damages, instead awarding a single sum for both. In these cases, for the purposes of the statistical analysis, the amount awarded was treated as aggravated damages rather than compensatory damages. This was on the basis that the court had made some award for aggravated damages and not none, and while aggravated damages are compensatory damages, compensatory damages are not aggravated damages.

Column Q: mode of trial (judge and jury, or judge alone). However, there was only one claim which involved a trial by judge and jury — that is, only one claim in which a jury decided not only that exemplary damages should be awarded but also the amount, and that was the final result in the claim.[90] On the basis that no meaningful analysis of the significance of the mode of trial could be done with only one such claim, this study (unlike the UK Study) did not conduct any statistical analysis for this variable.

Column R: notes. Any other pertinent information was recorded here.

So this study included some columns additional to those used in the UK Study: claim number; jurisdiction; defendant subcategory; and cause of action. It also omitted one column used in the UK Study: Rookes category, since the approach in that case has never been followed in Australia.

D Limitations

The authors of the UK Study identified three main limitations with their methodology. The first was sample size (146 claims in their study).[91] In this study, the sample size was significantly larger (253 claims). It is considered that this is sufficiently large to draw some meaningful conclusions. However, as with the UK Study, with respect to certain variables, the number of claims is sometimes small — for example, the number of claims falling within some of the combined cause of action categories. So there is a need for caution in interpreting those results. The second limitation identified by the authors of the UK Study was selection bias, as the sample comprised only decisions that could be accessed electronically.[92] In the UK Study, this had the effect of excluding most County Court decisions. However, in Australia, many lower court decisions are electronically available — for example, District Court (or equivalent) decisions in most jurisdictions. So this is a less significant issue here. The third limitation identified by the authors of the UK Study was that the data for some claims was accessed indirectly, via the decisions of appeal courts.[93] This limitation does not apply here as the study includes appeals, so all claims were accessed directly.

E Statistical Analysis

The statistical analysis was carried out by a statistician.[94] To enable useful comparison between Australia and the UK, the same type of analysis was conducted as in the UK Study. In short, the statistician performed Pearson’s chi-squared tests (for example, to test the association between the likelihood of being awarded exemplary damages and the cause of action category),[95] analysis of variance (‘ANOVA’) tests (for example, to test for significant differences in the amount of exemplary damages awarded between cause of action categories),[96] and linear regression analysis (for example, to test how well the amount of exemplary damages awarded predicts the amount of aggravated damages awarded).[97]

V Results

This section sets out the results of the statistical analysis of the sample.

A Overall

1 Across All Jurisdictions

Of the 253 claims, exemplary damages were awarded in 120 and not awarded in 133, giving a success rate of 47.43%. The mean award was $105,059.10, the median award $26,853.86, the minimum award $3,571.89, and the maximum award $4,167,202.14.[98] The distribution of exemplary damages awards is shown in Figure 1.

Figure 1: Distribution of Exemplary Damages Awards

The thick black line in the middle of the box denotes the median award. The edges of the box show the first and third quartiles.[99] The whiskers denote 1.5 times the interquartile range. Each dot represents an outlier between 1.5 and three times the interquartile range. And each star represents an extreme case. The maximum award of $4,167,202.14 is not shown.

Figure 2 is a temporal analysis, showing the number of awards of exemplary damages across all jurisdictions over the period of this study.

Figure 2: Number of Exemplary Damages Awards over Time

2 Jurisdiction by Jurisdiction

The number of claims for exemplary damages, and the success rate, for each Australian jurisdiction is shown in Table 1.

Table 1: Number of Claims for Exemplary Damages and Success Rates by Jurisdiction

Jurisdiction
Exemplary damages awarded
Total cases
% Awarded exemplary damages


No
Yes
Cth
8
15
23
65.22
ACT
6
0
6
0
NSW
55
53
108
49.07
NT
1
2
3
66.67
Qld
17
8
25
32
SA
13
13
26
50
Tas
4
2
6
33.33
Vic
14
18
32
56.25
WA
15
9
24
37.5
Total
133
120
253
47.43

By far the greatest number of claims were made in New South Wales (108 claims, 42.69%). In Victoria, with a comparable population, only 32 claims (12.65%) were made. The numbers of claims made in the Commonwealth, Queensland, Western Australia and South Australia were very similar, ranging from 23 to 25. The numbers of claims made in Tasmania, the Australian Capital Territory and the Northern Territory were also very similar, ranging from three to six. The highest success rate was in the Northern Territory, at 66.67% (only three claims were made and two succeeded) and then the Commonwealth, at 65.22%. The lowest success rate was in the Australian Capital Territory, where none of the six claims made were successful, then Queensland, at 32%. In New South Wales, where most claims were made, the success rate was 49.07%. However, statistical analysis suggests that there is no significant association between the jurisdiction and the likelihood of exemplary damages being awarded.[100]

The mean, median, minimum and maximum awards for each jurisdiction are shown in Table 2.

Table 2: Amount of Mean, Median, Minimum and Maximum Exemplary Damages Awards by Jurisdiction

Jurisdiction
Mean ($)
Median ($)
Minimum ($)
Maximum ($)
Cth
338,042.79
36,351.50
6,322.15
4,167,202.14
NSW
82,021.06
27,511.75
4,130.02
792,095.47
NT
44,018.81
44,018.81
44,018.81
44,018.81
Qld
175,159.56
49,285.94
14,081.70
595,912.90
SA
26,579.44
17,353.02
5,953.15
94,832.24
Tas
13,149.54
13,149.54
7,040.85
19,258.23
Vic
67,749.42
31,264.53
9,483.22
376,128.82
WA
12,077.63
10,194.86
3,571.89
35,204.24

The greatest mean award was in the Commonwealth, at $338,042.79,[101] then Queensland, at $175,159.56. The smallest mean award was in Western Australia, at $12,077.63, then Tasmania, at $13,149.54. In New South Wales, where most claims for exemplary damages were made, the mean award was $82,021.06. Similarly, the greatest median award was in Queensland, at $49,285.94, then the Northern Territory, at $44,018.81. The smallest median award was in Western Australia, at $10,194.86, then Tasmania, at $13,149.54. In New South Wales, the median award was $27,511.75. However, again, statistical analysis suggests that there is no significant association between the jurisdiction and the amount of exemplary damages awarded.[102]

B Cause of Action Category

Across all jurisdictions, the number of claims for exemplary damages in each category of cause of action is shown in Table 3. The greatest number of claims were made in the category of interference with the person: 99 claims (39.13%). Then was the category of defamation and privacy invasions: 49 claims (19.36%). The smallest number of claims were made in the category of interference with the person combined with interference with property: three claims (1.19%). Then was the category of interference with property combined with miscellaneous: four claims (1.58%).

Across all jurisdictions, the success rate for each category of cause of action is also shown in Table 3.

Table 3: Number of Claims for Exemplary Damages and Success Rates by Cause of Action

Cause of action category
Exemplary damages awarded
Total cases
% Awarded exemplary damages


No
Yes
0
Miscellaneous
10
7
17
41.18
1
Interference with the person
46
53
99
53.54
2
Interference with property
17
17
34
50
3
Defamation and privacy invasions
40
9
49
18.37
4
Abuse of power torts
2
13
15
86.67
5
Economic torts
8
11
19
57.89
6
Interference with person and abuse of power
4
4
8
50
7
Interference with person and property
0
3
3
100
8
Miscellaneous and interference with person
2
3
5
60
9
Miscellaneous and interference with property
4
0
4
0

The success rates are also shown graphically in Figure 3.

Figure 3: Success Rates by Cause of Action Category

The highest success rate was in the category of interference with the person combined with interference with property (column 7): all three claims succeeded. Then was the category of abuse of power torts (column 4): 86.67%. The lowest success rate was in the category of interference with property combined with miscellaneous: none of the four claims succeeded. Then was the category of defamation and privacy invasions (column 3): 18.37%. In the category of cause of action in which most claims for exemplary damages were made — interference with the person (column 1) — the success rate was 53.54%. Here, statistical analysis did show a significant association between the category of cause of action and the likelihood of exemplary damages being awarded.[103]

Across all jurisdictions, the mean, median, minimum and maximum awards for each category of cause of action are shown in Table 4.

Table 4: Amount of Mean, Median, Minimum and Maximum Exemplary Damages Awards by Cause of Action Category

Category
Mean ($)
Median ($)
Minimum ($)
Maximum ($)
0
106,247.57
57,843.39
3,571.89
376,128.82
1
51,875.21
20,650.10
4,130.02
595,912.90
2
280,292.19
26,201.98
5,162.53
4,167,202.14
3
32,010.01
19,258.23
7,040.85
137,592.82
4
101,582.08
26,853.86
6,322.15
792,095.47
5
165,071.96
103,194.62
5,359.45
578,659.73
6
37,179.63
21,845.70
10,194.86
94,832.24
7
162,534.19
155,543.90
145,405.99
186,652.68
8
96,078.79
49,285.94
49,285.94
189,664.49

The greatest mean award was in the category of interference with property (category 2), at $280,292.19. This was affected by the extreme outlier claim, a passing off case. Then was the category of economic torts (category 5), at $165,071.96. The smallest mean award was in the category of abuse of power torts (category 3), at $32,010.01. This was followed by the category of interference with the person combined with abuse of power torts (category 6), at $37,179.63. In the category of cause of action in which most claims were made, interference with the person (category 1), the mean award of exemplary damages was $51,875.21. As to median awards, the greatest was in the category of interference with the person combined with interference with property (category 7), at $155,543.90. Then was the category of economic torts (category 5), at $103,194.62. The smallest median was in the category of defamation and privacy invasions (category 3), at $19,258.23, followed by the category of interference with the person (category 1), at $20,650.10. However, statistical analysis shows no significant association between the category of cause of action and the amount of exemplary damages awarded.[104]

Figures 4 and 5 present a temporal analysis of selected categories of cause of action, namely: (i) defamation and privacy invasions; and (ii) miscellaneous, and miscellaneous combined with, first, interference with the person and, second, interference with property. The first category captures all claims for exemplary damages for defamation. The other categories capture all claims for exemplary damages for negligence. As seen earlier,[105] defamation and negligence are two key areas in which, during the period covered by this study, there has been significant statutory intervention in Australia, limiting or excluding the availability of exemplary damages. Figure 4 shows the number of awards of exemplary damages for defamation across all jurisdictions over the period of this study.

Figure 4: Number of Awards for Exemplary Damages for Defamation

Figure 5 does the same for negligence.

Figure 5: Number of Awards for Exemplary Damages for Negligence

C Defendant Category

Across all jurisdictions, the number of claims for each category of defendant is shown in Table 5. The greatest number of claims were made in the category of public body: 102 claims (40.32%). Then was the category of individual: 88 claims (34.78%). And finally, the category of corporation: 63 claims (24.9%). Across all jurisdictions, the success rate for each category of defendant is also shown in Table 5.

Table 5: Number of Claims and Success Rates by Defendant Category

Defendant category
Exemplary damages awarded
Total cases
% Awarded exemplary damages


No
Yes
Public body
43
59
102
57.84
Corporation
43
20
63
31.75
Individual
47
41
88
46.59
Total
133
120
253
47.43

The success rates are shown graphically in Figure 6.

Figure 6: Success Rates by Defendant Category

The same ranking appears. The highest success rate was in the category of public body (column 1): 57.84%. Then the category of individual (column 3): 46.59%. And finally the category of corporation (column 2): 31.75%. Statistical analysis shows a significant association between the category of defendant and the likelihood of being awarded exemplary damages.[106]

Across all jurisdictions, the mean, median, minimum and maximum awards for each category of defendant are shown in Table 6.

Table 6: Amount of Mean, Median, Minimum and Maximum Exemplary Damages Awards by Defendant Category

Category
Mean ($)
Median ($)
Minimum ($)
Maximum ($)
0
65,414.89
26,797.23
5,162.53
792,095.47
1
326,210.36
43,988.38
7,233.25
4,167,202.14
2
54,229.43
21,483.09
3,571.89
433,994.80

Here, a different ranking emerges. The greatest mean award was in the category of corporation (category 1): $326,210.36. This was affected by the extreme outlier claim. Then was the category of public body (category 0): $65,414.89. And finally, the category of individual (category 2): $54,229.43. Similarly, the greatest median award was in the category of corporation (category 1): $43,988.38. Then the category of public body (category 0): $26,797.23. And finally, the category of individual (category 2): $21,483.09. Statistical analysis again shows a significant association between the category of defendant and the amount of the exemplary damages award.[107]

D Defendant Subcategory

As it became apparent from an early stage of the case review that there were a large number of claims against police, a separate analysis of these claims was performed. Across all jurisdictions, 78 claims were made against police (being 30.83% of all claims). Moreover, the success rate in those claims was 54/78 (69.23%). Statistical analysis shows a significant difference in success rates of claims against police as compared to those against other categories of defendant.[108]

Across all jurisdictions, the mean award in claims against police was $62,471.14 and the median award was $26,144.25. But this time, statistical analysis showed no significant difference in the amount of exemplary damages awards against police as compared to those against other categories of defendant.[109]

E Other Types of Damages

1 Aggravated Damages

Across all jurisdictions, both exemplary damages and aggravated damages were awarded in 63 claims. Aggravated damages alone were awarded in 43 claims. And exemplary damages without aggravated damages (ie exemplary damages and compensatory damages) were awarded in 57 claims. In other words, when exemplary damages were awarded, aggravated damages were also awarded in 63/120 claims (52.5% of the time) and not awarded in 57/120 claims (47.5% of the time). When exemplary damages were not awarded, aggravated damages were awarded in 43/133 claims (32.33% of the time) and not awarded in 90/133 claims (67.67% of the time). Statistical analysis shows a significant association between the likelihood of being awarded exemplary damages and the likelihood of being awarded aggravated damages.[110]

Across all jurisdictions, where both exemplary and aggravated damages were awarded (as seen, 63 claims), the mean award of exemplary damages was $57,878.98 and the mean award of aggravated damages was $38,367.99. The median award of exemplary damages was $26,604.25 and the median award of aggravated damages was $23,474.97. But statistical analysis shows no significant association between the amount of exemplary damages awarded and the amount of aggravated damages awarded.[111]

2 Compensatory Damages

Across all jurisdictions, both exemplary damages and compensatory damages (excluding aggravated damages)[112] were awarded in 52 claims. Compensatory damages without exemplary damages were awarded in 74 claims. And exemplary damages without compensatory damages were awarded in five claims. In other words, when exemplary damages were awarded, compensatory damages were also awarded in 100/120 claims (83.33% of the time) and not awarded in 20/120 claims (16.67% of the time). When exemplary damages were not awarded, compensatory damages were awarded in 98/133 claims (73.68% of the time) and not awarded in 35/133 claims (26.32% of the time). But statistical analysis shows no significant association between the likelihood of being awarded exemplary damages and the likelihood of being awarded compensatory damages.[113]

Further, across all jurisdictions, where both exemplary and compensatory damages were awarded (as seen, 52 claims) the mean award of exemplary damages was $165,536.85 and the mean award of compensatory damages was $550,872.13. The median award of exemplary damages was $30,703.19 and the median award of compensatory damages was $98,807.10. Here, statistical analysis shows a significant association between the amount of exemplary damages awarded and the amount of compensatory damages awarded.[114]

VI Analysis

This part analyses the results of the study in light of the concerns frequently expressed by courts and commentators about exemplary damages in Australia, and the various statutory interventions. The part is divided into two sections. The first considers the results of the study standing alone. The second considers the results in comparison with those of the UK Study.

A Results Standing Alone

1 Rate at Which Exemplary Damages Are Claimed and Awarded

Plainly, exemplary damages are not extinct in Australia. On one view, they are not even especially rare or exceptional, as commonly asserted.[115] Over the period of the study, exemplary damages were sought in 253 claims, and the success rate in those claims was 47.43%. In other words, they were claimed somewhere in Australia more than once a month, and about half the time they were awarded. However, as the authors of the UK Study note, the success rates yielded by this study would have been lower if the study had included claims in which the court found against the plaintiff on liability but proceeded to consider exemplary damages and to make a putative award in case it was wrong.[116]

Further, seen in broader context, it seems likely that the number of claims per capita and the success rate have declined over time, given the various statutory interventions limiting or excluding the availability of exemplary damages in Australia. This trend is not readily apparent from Figure 2 above,[117] which graphs the number of awards of exemplary damages over time, for the period of this study. But it would probably emerge from the same graph over a much longer time period. Further empirical research would need to be done — for example, counting the number of claims for exemplary damages per capita and the success rate in those claims since 1920, when the decision of the High Court in Whitfeld recognised exemplary damages as part of Australian law. This would be a significantly larger project than this study, requiring far greater resources and funding.

Further, even over the period of this study, seen in the context of the number of claims for damages generally, the number of claims for exemplary damages is likely to be small. For example, in one year (2016) in one jurisdiction (New South Wales) in one court (Supreme Court) and in one division of that Court (Common Law Division), personal injury claims[118] disposed of by themselves numbered 291.[119]

2 Amount of Exemplary Damages Awarded

As seen, Australian courts frequently emphasise the need for restraint and moderation in the assessment of exemplary damages.[120] This may be borne out of a fear that exemplary damages awards in this country will follow the example of the United States, where massive awards have been made. For example, in 2000 a Florida jury awarded US$145 billion in punitive damages to a class of plaintiffs in a claim against a tobacco manufacturer.[121] However, across all Australian jurisdictions, this study yielded a mean exemplary damages award of $105,059.10 and an even more modest median award of $26,853.86. What is more, the mean award was significantly affected by one very large award in a Federal Court passing off case.[122] In that case, the applicant owned intellectual property rights in relation to ugg boots. The respondent manufactured and marketed counterfeit ugg boots. The applicant brought a common law claim for passing off but also relied on statutory causes of action, including breach of copyright and contraventions of the Trade Practices Act 1974 (Cth). The Court dealt with common law exemplary damages for passing off and additional damages under s 115(4) of the Copyright Act 1968 (Cth) together, drawing no distinction between them. The award of over $4 million in exemplary damages is an extreme outlier in this study. Figure 1 above shows that most cases fall within the range of $0 to $150,000.[123] This is reflected in the median award, which is approximately four times lower than the mean.

Further, looking at individual jurisdictions, the Commonwealth is the exception, where the mean award of exemplary damages was $338,042.79 (given the effect of the extreme outlier claim). Otherwise, the mean and median awards were generally less than $100,000. Indeed, in four jurisdictions (Western Australia, Tasmania, South Australia and the Northern Territory), the mean and median awards were less than $50,000.

It is plain, therefore, that Australian courts have not, so far, followed the US example. But could it still be said that awards of exemplary damages in this country are excessive? As the authors of the UK Study note, we need a reference point to decide.[124] They identify three possibilities. The first is compensatory damages awards. In this study, across all jurisdictions, where both exemplary damages and compensatory damages were awarded, the mean award of compensatory damages ($550,872.13) was approximately three times the mean award of exemplary damages ($165,536.85). The same holds for the median awards of exemplary and compensatory damages (respectively, $98,807.10 and $30,703.19). A second possible reference point is average incomes. In Australia, the average annual income of a full-time employee in 2000 (the first year of this study) was $54,043.73 (adjusted for inflation to 2017).[125] By 2016 (the last year of this study) it was roughly $84,418.37.[126] Seen in light of these numbers, the mean and especially the median awards of exemplary damages do not seem particularly high. A third possible reference point is the position in other jurisdictions. The UK position is discussed in the next part of this article.[127] In the US, a study in 2011 yielded a mean award of US$2,400,000 and a median award of US$103,500 (both adjusted for inflation).[128] So awards of exemplary damages in the US are significantly higher than in Australia.

These reference points suggest that exemplary damages awards in Australia are (generally) not excessive.

As to the predictability of awards, a commonly expressed objection to exemplary damages is that they are subjective and discretionary, and therefore uncertain and unpredictable, both as to whether or not they will be awarded and, where they are, as to their amount.[129] But this objection does not appear to hold in light of this study. Admittedly, the exemplary damages awards made in the claims the subject of the study ranged from $3,571.89 to $4,167,202.14, with a standard deviation of $392,901.52. But this range was significantly affected by the extreme outlier claim. Further, looking at each jurisdiction separately, a much more uniform picture emerges, especially in the states and territories. In all but two jurisdictions (excluding the Commonwealth, where the extreme outlier claim was made), the range between the minimum award of exemplary damages and the maximum award was less than $100,000.[130] Indeed, in three of these jurisdictions, the range was less than $50,000.[131] So in these jurisdictions at least, this objection does not appear to hold.

3 Cause of Action

As seen, under Australian law, with some exceptions,[132] exemplary damages are available for any cause of action. Australia never adopted the ‘cause of action test’ applied in England between 1964 and 2001, according to which exemplary damages were only available for causes of action for which they had been available prior to the decision of the House of Lords in Rookes.[133] So as a matter of law, in Australia, the particular cause of action does not generally matter.

In practice, however, given the test for exemplary damages applied in Australia, requiring ‘conscious wrongdoing in contumelious disregard’ of the plaintiff’s rights,[134] the heartland for exemplary damages is undoubtedly the intentional torts. This is borne out by the results of this study. The number of claims falling within the miscellaneous category, which included negligence and other unintentional torts, was only 17/253 (6.72%). Moreover, the study demonstrates a statistically significant association between the cause of action category and the likelihood of being awarded exemplary damages.[135]

There is some overlap between the categories of cause of action in which the greatest number of awards of exemplary damages were made (the category of interference with the person, and the category of defamation and privacy invasions) and the categories of cause of action with the highest rate of success (the categories of interference with the person combined with interference with property, abuse of power torts, and economic torts, closely followed by the categories of interference with the person and interference with property). With the exception of the category of defamation and privacy invasions, claims for exemplary damages in these categories of cause of action are unaffected by the statutory intervention excluding or limiting the availability of exemplary damages discussed earlier.[136] The two categories of cause of action most affected by the statutory intervention are defamation and privacy invasions, and miscellaneous (including, in particular, negligence claims resulting in death or personal injury).

In relation to defamation, unsurprisingly, the study shows a sharp drop-off of awards of exemplary damages in the years following the introduction of the uniform defamation Acts, abolishing exemplary damages in defamation actions across Australia. Figure 4 shows that in the period 2000 to 2007, there were seven such awards but thereafter none until 2012 (one claim)[137] and 2015 (one claim).[138] One would expect there to be few, if any, further such awards, as the time since the enactment of the 2005 Acts increases. But again, it would be interesting to undertake further empirical research counting exemplary damages awards in defamation claims over a longer period ending in 2005. What is more, across the period of this study, the success rate in defamation claims was just 18.37%, the lowest of all categories of cause of action save for the very small category of miscellaneous combined with interference with property, in which no claims were successful. The mean award of $32,010.01 for defamation claims was also the lowest of all categories of cause of action. These results do not indicate any serious problem with exemplary damages awards for defamation.

In relation to negligence, a declining trend is less obvious, as is shown in Figure 5. However, since the enactment of the civil liability legislation in 2003 and 2004 in several Australian states and territories and, in particular, provisions in that legislation abolishing exemplary damages in personal injury actions, only two further awards of exemplary damages for negligence were made in those jurisdictions.[139] Again, it would be useful to conduct further empirical research looking at the rate of exemplary damages awards in claims for negligence resulting in personal injury over a longer period of time leading up to the enactment of the civil liability legislation. Such a study would be likely to demonstrate that exemplary damages have always been a marginal remedy in negligence claims.

So in both these categories of cause of action, the study demonstrates that the statutory intervention has achieved its purpose. But it does not answer whether the mischief that the legislation was designed to remove was real or significant.

4 Defendant

The study demonstrates a statistically significant association between the category of defendant and the likelihood of being awarded exemplary damages.[140] The greatest number of claims for exemplary damages and the highest success rate both occurred in the same category of defendant — public bodies.[141] As will be seen,[142] this is due in no small measure to the relatively large number of claims brought against police, and the relatively high success rates in those claims.

Moreover, the study also demonstrates a significant association between the category of defendant and the amount of exemplary damages awarded.[143] The largest overall awards (mean and median) were made against corporations, then public bodies and then individuals.[144] The mean award against corporations was affected by the extreme outlier claim. Generally, corporations and public bodies might be expected to have deeper pockets than individuals. Under Australian law, the greater the financial means of the defendant, the larger the award of exemplary damages should be.[145] This is so as to best achieve the punishment and deterrence goals of the remedy.[146] The results of the study are consistent with this principle and suggest it is applied by Australian courts.

5 Police

Perhaps the most striking result to come out of this study concerns claims against police. As mentioned, there is a commonly held view that exemplary damages today are most frequently awarded against police. This view may be unfounded in the UK.[147] But in Australia, it is strongly supported by this study. A relatively large number of claims were brought against police: across all jurisdictions, 78 of 253 claims (30.83%).[148] In other words, in Australia today, about one in every three claims for exemplary damages is brought against police. What is more, these claims are made almost exclusively in two categories of cause of action: the category of interference with the person (principally assault, false imprisonment and wrongful arrest) and the category of abuse of power (principally malicious prosecution).[149] The success rates against police are also significantly higher than in exemplary damages claims generally. Of the 78 claims against police, exemplary damages were awarded in 54 (69.23%).[150] Recall that the overall success rate in exemplary damages claims was just under 50%. There is a statistically significant difference between the success rates against police and those against other defendants.

The study also demonstrates that the majority of claims for exemplary damages against police are brought in New South Wales (41 of 78 claims, 52.56%). Further study, empirical and otherwise, is needed to identify the reasons for this. Plainly, it is not just a function of the larger population in New South Wales. In Victoria, with a comparable population, only 10 claims for exemplary damages against police were made. Part of the explanation for the relatively fewer claims against police outside New South Wales must be the legislation in place in several jurisdictions restricting the vicarious liability of the state for conduct of police officers — in particular, providing that such liability does not extend to exemplary damages.[151] There is no such legislation in New South Wales. In these jurisdictions, claims for exemplary damages against police must be made against individual officers, who might be expected to have fewer resources and therefore be not worth pursuing.

Looking at the amount of awards of exemplary damages against police, across all jurisdictions, the mean award was $62,471.14 and the median award was $26,144.25.[152] So the mean award was just over half the overall mean of $105,059.10 (but, unlike the overall mean award, was not affected by the extreme outlier claim). And the median award was almost the same as the overall median of $26,853.86. The maximum award of exemplary damages made against police was $792,095.47. Unsurprisingly, therefore, there was no statistically significant difference between the amount of exemplary damages awarded against police and that awarded against other defendants. In other words, in assessing exemplary damages awards against police, courts in Australia exercise the same restraint as they do in other claims for exemplary damages.

6 Other Types of Damages

As seen, Australian courts have said that there is ‘no necessary proportionality’ between an award of exemplary damages and an award of compensatory (including aggravated) damages.[153] But they have also accepted, as recognised by Lord Devlin in Rookes,[154] that the amount of compensatory damages awarded is relevant to the decision to award exemplary damages at all and, if so, how much. This is because the amount of compensation awarded may be adequate to punish the defendant and to achieve the goals of an exemplary damages award; only if it is not does the question of exemplary damages arise.[155] The results of this study are not unambiguously consistent with these principles.

As to aggravated damages and their relationship (if any) with exemplary damages, there is a statistically significant association between the likelihood of being awarded exemplary damages and the likelihood of being awarded aggravated damages. When courts award exemplary damages they often award aggravated damages too.[156] However, where both exemplary and aggravated damages are awarded, there is no statistically significant association between the amount of one and the amount of the other. So courts must be taking into account different factors when assessing awards of exemplary and aggravated damages in a particular case.

As to the relationship (if any) between exemplary damages and general compensatory damages, the reverse appears to be true. There is no statistically significant association between the likelihood of being awarded exemplary damages and the likelihood of being awarded compensatory damages. Perhaps this is unsurprising. The fact that the plaintiff has suffered loss (which may or may not justify an award of aggravated damages) does not tell us much about whether an award of exemplary damages is appropriate. However, where both are awarded, there is a statistically significant association between the amount of exemplary damages awarded and the amount of compensatory damages awarded. So it may be that Australian courts do take into account the amount of compensation available when assessing exemplary damages. In other words, the size of the plaintiff’s loss (at least, the loss compensated by an award of general, rather than aggravated damages) may have some bearing on the amount of exemplary damages awarded.

The study also permits some observations in relation to the practice of Australian courts in making combined (global) awards of exemplary, aggravated and/or compensatory damages versus separate (itemised) awards of the same. It appears that it is relatively rare for courts in Australia to make combined awards. Of the 253 claims, in only seven were combined awards of exemplary, aggravated and/or compensatory damages made.[157] This is consistent with recent authority. In New South Wales v Cuthbertson,[158] the New South Wales Court of Appeal held that, although it is not an error for a court to make a single award for exemplary and aggravated damages (following the High Court in New South Wales v Ibbett),[159] this practice was not to be encouraged.[160] This was because (inter alia) it gave rise to difficulties on appeal, and because an award of exemplary damages was less likely to have a punitive or deterrent effect if it was not separately identified.[161]

However, it is not uncommon for Australian courts, while making a separate (if any) award of exemplary damages, to make a combined award of aggravated and compensatory damages. Of the 253 claims, in 32 a single sum was awarded for both these heads of damages.[162] There were, however, some jurisdictional differences, with courts in New South Wales and Western Australia generally more likely than those in other jurisdictions to adopt this practice.

7 Multiple Parties and Causes of Action

The study also permits some observations on the practice of Australian courts in claims where there are multiple plaintiffs, defendants or causes of action made out. Each is considered below, starting with defendants.

Under English law, where there are two or more defendants, a single award (if any) of exemplary damages against all should generally be made.[163] That award cannot exceed the amount appropriate against the least culpable defendant. So if against one or more of the defendants no award of exemplary damages is appropriate, no award of exemplary damages at all can be made, even if it would otherwise be justified against the other defendants.[164] This approach is open to the objection that, in these circumstances, the more culpable defendants, against whom exemplary damages were justified, avoid liability for this remedy, with the consequence that the goals of punishment and deterrence fail.[165]

Australian courts do not always adopt this practice. In this study, examples can be found in most jurisdictions of courts making separate awards against multiple defendants.[166] So it appears that Australian courts prefer to treat the liability of multiple defendants for exemplary damages as several, rather than joint.[167]

As to multiple plaintiffs, under English law, where there are two or more plaintiffs, courts generally assess the total amount of exemplary damages justified by the defendant’s conduct, then divide that amount by the number of plaintiffs.[168] The rationale for this approach is that otherwise there is a danger of over-punishment of the defendant.[169] However, the approach leads to difficulties, especially where not all plaintiffs are before the court at the same time.[170]

Australian courts do not always adopt this practice, either. In the study, examples can be found in several jurisdictions of courts making a separate award of exemplary damages for each of several plaintiffs. These awards are not assessed by first determining the maximum overall award of exemplary damages and then apportioning that between the plaintiffs.[171] It appears that, as a general rule, Australian courts do not follow the practice laid down in the UK and separately consider each plaintiff’s claim on its merits.

As to multiple causes of action, it is also not uncommon for Australian courts to make separate awards of exemplary damages in respect of each separate cause of action made out by the plaintiff, rather than an aggregate award for all causes of action. Examples of this practice can be found in several jurisdictions.[172] The general approach of Australian courts may be summed up by the New South Wales Court of Appeal in Adams v Kennedy:

[A]lthough strictly it would be proper to award a separate amount for each cause of action, it seems to me that since the different causes of action arose out of the one series of closely connected events, it is appropriate to award one aggregate figure in respect of all the causes of action.[173]

B Results in Comparison with the UK Study

The consistency of approach to methodology allows for meaningful comparative analysis of the results of this study and those of the UK Study.

Generally, exemplary damages are in more robust health in Australia than in the UK. Over a very similar period of time,[174] the number of claims was greater in Australia: 253 claims compared to 146 claims in the UK Study.[175] The success rate in those claims was also higher in Australia: 120/253 (47.43%) compared to 58/146 (39.72%) in the UK Study.[176] The amount of exemplary damages awarded was also greater in Australia: $105,059.10 overall mean award and $26,853.86 overall median award, compared to £12,625 (approximately $23,275.13)[177] mean award and £7,630 (approximately $14,063.09) median award in the UK Study.[178] So the Australian mean is about 4.5 times greater than the UK mean, and even the median is almost twice as large in Australia. In short, exemplary damages are claimed more often, and awarded more often and in greater amounts, in Australia than in the UK. This is perhaps not surprising, notwithstanding the significant statutory intervention in Australia limiting or excluding the availability of exemplary damages. Outside of claims affected by that legislation, the Australian approach to exemplary damages is much less restrictive than the ‘parsimonious’[179] Rookes approach still followed in the UK.

In relation to causes of action, the categories of cause of action in which exemplary damages are most often claimed and awarded are not the same in Australia and the UK. As seen,[180] in Australia exemplary damages are most often claimed in the categories of interference with the person (99 claims), defamation and privacy invasions (49 claims), and interference with property (34 claims). In the UK, they are most often claimed in the categories of interference with property (52 claims), interference with the person (39 claims), and the economic torts (29 claims).[181] The number of defamation claims in the UK (eight) was much lower than in Australia, whereas the number of economic tort claims was much lower in Australia than in the UK, constituting 7.51% of claims compared to 19.8% of claims respectively.[182]

There was some overlap between the categories of cause of action which produced the highest success rates in Australia and the UK. In Australia, the highest success rates were in the category of interference with the person combined with interference with property (100%), abuse of power torts (86.67%), and the economic torts (57.89%), closely followed by interference with the person (53.54%). In the UK, the highest success rates were in the economic tort claims (55.2%), interference with property (53.8%), then abuse of power torts (40%).[183] In particular, in the UK, the high success rate in the economic torts category was exclusively due to insurance fraud cases, framed in deceit.[184] Such cases did not figure in this study. One reason for this difference must be the existence in Australia of s 18 of the Australian Consumer Law[185] (and its state and territory equivalents). Such claims may be more likely to be brought on the basis of this statutory cause of action than the common law action for deceit, and to seek the remedies available under the statute, rather than common law remedies including exemplary damages. Further, in the UK, a significant proportion of the interference with property claims were landlord and tenant disputes. These did not really figure in this study either. Again, applicable statutory regimes in Australia may go some way to explaining this difference. No doubt, the Rookes approach is also relevant in explaining the cause of action differences between Australia and the UK.

In relation to defendants, there were some differences between Australia and the UK. As to the number of claims made, as seen,[186] the ranking of categories of defendant from most to least number of claims was public bodies, then individuals and then corporations. In the UK it was the same.[187] As to success rates, the ranking was the same again in Australia, but in the UK it was individuals, then corporations and then public bodies. So in Australia, most successful claims were brought against public bodies, but in the UK, even though most claims were brought against public bodies, that was the category of defendant in which the least claims were successful. This is somewhat surprising, given that one of the Rookes categories is (particular) claims against public bodies.[188] As to the amount of exemplary damages awarded, in Australia the ranking from greatest to least awards was corporations, then public bodies and then individuals. It was the same in the UK.[189] As the authors of the UK Study note, in the UK, as in Australia, this ‘can arguably be explained on the ground that ... [corporations and public bodies] are likely to be better resourced’ and ‘it is well established that the defendant’s wealth is relevant to the assessment of ... [exemplary] damages’.[190]

Finally, in relation to exemplary damages and other types of damages, as seen,[191] in Australia there is a statistically significant association between the likelihood of being awarded exemplary damages and the likelihood of being awarded aggravated damages. But there is no association between the amount of exemplary damages and the amount of aggravated damages, where both are awarded. Further, there is no association between the likelihood of being awarded exemplary damages and the likelihood of being awarded compensatory damages. But there is a statistically significant association between the amount of exemplary damages and the amount of compensatory damages, where both are awarded. In the UK, the relationship between exemplary and aggravated damages is somewhat different. As in Australia, there is a probable association between the likelihood of an award of exemplary damages and the likelihood of an award of aggravated damages. But there is also a (possible, loose) relationship in relation to the amount awarded.[192] As in Australia, so too in the UK there is a (possible, loose) association between the amount of exemplary damages and the amount of compensatory damages, where both are awarded.[193] So in both jurisdictions, it is possible that the extent of the plaintiff’s loss plays some role in explaining the amount of exemplary damages awarded.[194]

VII Conclusion

Inspired by the UK Study, this study answered the call of various commentators and courts for an empirical study of exemplary damages in Australia. Its key findings are as follows. Exemplary damages are not extinct in Australia: they are alive and well. Across all Australian jurisdictions, over the period of this study, exemplary damages were sought in 253 claims and awarded in 120 of those claims, yielding a success rate of just less than 50% (47.43%). The mean award was $105,059.10 and the median award was $26,853.86. Looking at the states and territories, by far the greatest number of claims were brought in New South Wales (108 claims, or 42.69%). The highest success rate was in the Northern Territory (two out of three claims, or 66.67%), and the greatest mean award was in the Commonwealth ($338,042.79).

In relation to categories of cause of action, across all jurisdictions, exemplary damages were most often sought in claims involving interference with the person, especially assault and false imprisonment. They were most often awarded where the claim included both interference with the person and interference with property — for example, trespass. The greatest awards, on average, were also made in the interference with property claims — for example, common law intellectual property cases. The number of defamation claims dropped off steeply during the period of the study, and both the success rates and the amounts of awards were among the lowest in this category of claims.

In relation to categories of defendant, across all jurisdictions, the greatest number of claims for exemplary damages and the highest success rates were against public bodies. But, on average, larger awards were made against corporations.

More than three quarters of the claims made against public bodies were brought against police. Further, the success rate against police was significantly higher than the overall success rate, and the success rates against any other category of defendant. However, awards of exemplary damages against police were lower, on average, than the overall mean award, and lower than against any other category of defendant, except individuals. So it appears that Australian courts are more willing to award exemplary damages against police than other defendants, but exercise restraint when assessing those awards.

Finally, in general terms, it appears from the study that exemplary damages tend to go hand in hand with aggravated damages. And when assessing exemplary damages, courts do take into account the amount of compensatory damages awarded to the plaintiff.

The concerns frequently expressed by courts and commentators about exemplary damages in Australia, including as to their overuse and excessive awards, should be considered afresh in light of these findings. It may be that the concerns have been overstated. It is also hoped that the results of the study will be useful in assessing the effect of the various statutory interventions in Australia limiting or excluding the availability of exemplary damages, and at the policy formulation stage, should it be thought that further statutory intervention is desirable.

VIII Appendix: Key Legislative Provisions Limiting or Excluding the Availability of Exemplary Damages

A Claims Surviving for the Benefit of the Estate after the Death of the Plaintiff

ACT
(1) This section applies to a cause of action that survives under this part for the benefit of a dead person’s estate.
(2) The damages recoverable do not include exemplary damages.
NSW
(2) Where a cause of action survives as aforesaid for the benefit of the estate of a deceased person, the damages recoverable for the benefit of the estate of that person:
(a) shall not include:
(i) any exemplary damages ...
NT
(1) Where a cause of action survives under this Part for the benefit of the estate of a deceased person, the damages recoverable for the benefit of the estate of that person:
(a) do not include exemplary damages ...
Qld
(2) Where a cause of action survives pursuant to subsection (1) for the benefit of the estate of a deceased person, the damages recoverable in any action brought —
...
(b) shall not include exemplary damages ...
SA
(1) Where a cause of action survives as aforesaid for the benefit of the estate of a deceased person, the damages recoverable for the benefit of the estate of that person —
...
(b) shall not include any exemplary damages ...
Tas
(3) Where a cause of action survives under this section for the benefit of the estate of a deceased person, the damages recoverable for the benefit of that estate —
(a) shall not include exemplary damages ...
Vic
(2) Where a cause of action survives as aforesaid for the benefit of the estate of a deceased person the damages recoverable for the benefit of the estate of that person —
(a) shall not include any exemplary damages ...
WA
(2) Where a cause of action survives as aforesaid for the benefit of the estate of a deceased person, the damages recoverable for the benefit of the estate of that person —
(a) shall not include any exemplary damages ...

B Defamation

ACT
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
NSW
(3) In particular, damages for defamation:
(a) shall not include exemplary damages ...
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
NT
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
Qld
A plaintiff can not be awarded exemplary or punitive damages for defamation.
SA
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
Tas
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
Vic
A plaintiff cannot be awarded exemplary or punitive damages for defamation.
WA
A plaintiff cannot be awarded exemplary or punitive damages for defamation.

C Personal Injury: General

Cth
(previously Trade Practices Act 1974 (Cth) s 87ZB)
(1) A court must not, in a proceeding to which this Part applies, award exemplary damages or aggravated damages in respect of death or personal injury.
(2) This section does not affect whether a court has power to award exemplary damages or aggravated damages:
(a) otherwise than in respect of death or personal injury; or
(b) in a proceeding other than a proceeding to which this Part applies.
NSW
In an action for the award of personal injury damages where the act or omission that caused the injury or death was negligence, a court cannot award exemplary or punitive damages or damages in the nature of aggravated damages.
NT
A court must not award aggravated damages or exemplary damages in respect of a personal injury.
Qld
(1) A court can not award exemplary, punitive or aggravated damages in relation to a claim for personal injury damages.
(2) Subsection (1) does not apply to a claim for personal injury damages if the act that caused the personal injury was —
(a) an unlawful intentional act done with intent to cause personal injury; or
(b) an unlawful sexual assault or other unlawful sexual misconduct.

D Personal Injury: Motor Vehicle

ACT
Road Transport (Third-Party Insurance) Act 2008 (ACT) s 22(1)(e)
(1) A CTP policy does not insure against the risk of 1 or more of the following:
...
(e) liability to pay exemplary, punitive or aggravated damages ...
NSW
A court shall not award exemplary or punitive damages to a person in respect of a motor accident.
A court cannot award exemplary or punitive damages to a person in respect of a motor accident.
Exemplary or punitive damages cannot be awarded to a person in respect of a motor accident.
NT
(1) Subject to subsections (2) and (3), the Commission must indemnify the owner or driver of a Territory motor vehicle for any relevant liability incurred in respect of the death of, or injury to, a person arising from a motor accident caused by, or arising out the use of, the motor vehicle outside the Territory but within Australia.
(2) The Commission is not bound to indemnify a person under subsection (1) where that person is already indemnified under any contract of insurance or under the law applicable in the place where the accident occurred.
(3) A relevant liability is:
(a) a liability for damages (but not for exemplary or punitive damages) ...
Qld
(3) This policy does not insure against a liability to pay exemplary, punitive or aggravated damages.
SA
An insurer is not liable to pay any aggravated damages or exemplary or punitive damages awarded in an action against the insured person in respect of death or bodily injury caused by or arising out of the use of a motor vehicle insured under this Part and the insured person is not entitled to be indemnified by the insurer in respect of such an award.
Tas
(1) By virtue of this Act, but subject to and in accordance with the provisions thereof, the Board is bound to indemnify an owner or user of a motor vehicle, or his legal personal representatives, in respect of any liability (not being a contractual liability) incurred by him in respect of the personal injury to a person resulting directly from a motor accident involving that motor vehicle in this State on or after the appointed day.
...
(6) For the purposes of this Part, the Board is not bound to pay exemplary or punitive damages claimed against —
(a) an indemnifiable person; or
(b) the Board.
Vic
Transport Accident Act 1986 (Vic) s 93(7) (citations omitted)
(7) Damages of any kind in respect of an injury cannot be recovered in proceedings in accordance with sub-sections (2), (3) and (4) other than damages —
(a) for pecuniary loss but only if —
(i) the assessment of damages before any reduction in respect of the person’s responsibility for the injury is more than $30,520 but less than $686,840, in which case the amount that can be recovered is that amount so assessed as reduced first under sub-section (11) and secondly in respect of the person’s responsibility for the injury; or
(ii) the assessment of damages before any reduction in respect of the person’s responsibility for the injury is more than $686,840, in which case the amount that can be recovered is $686,840 as reduced first under sub-section (11) and secondly in respect of the person’s responsibility for the injury;
(b) for pain and suffering but only if —
(i) the assessment of damages before any reduction in respect of the person’s responsibility for the injury is more than $30,520 but less than $305,250, in which case the amount that can be recovered is that amount so assessed as reduced first under sub-section (11) and secondly in respect of the person’s responsibility for the injury; or
(ii) the assessment of damages before any reduction in respect of the person’s responsibility for the injury is more than $305,250, in which case the amount that can be recovered is $305,250 as reduced first under sub-section (11) and secondly in respect of the person’s responsibility for the injury.
(To similar effect)

E Personal Injury: Workers’ Compensation

NSW
A court may not award exemplary or punitive damages to a person in an award of damages.
Qld
Workers’ Compensation and Rehabilitation Act 2003 (Qld) s 306B, (previously WorkCover Queensland Act 1996 (Qld) s 319)
(1) A court can not award exemplary or punitive damages against WorkCover in a claimant’s proceeding for damages.
(2) However, the court may give a separate judgment against an employer for the payment of exemplary or punitive damages if the court considers that the employer’s conduct is so reprehensible that an award of exemplary or punitive damages is justified.
(3) WorkCover can not indemnify an employer against an award of exemplary or punitive damages.
SA
A court may not award exemplary or punitive damages to a person in an award of damages to which this Part applies.
Vic
(22) A court must not, in proceedings in accordance with this section, award to a worker in respect of an injury —
(a) pecuniary loss damages ... ; or
(b) pain and suffering damages ... ; or
(c) damages of any other kind, other than damages in the nature of interest.

F Police

Cth
(1) The Commonwealth is liable in respect of a tort committed by a member or a protective service officer in the performance or purported performance of his or her duties as such a member or a protective service officer in like manner as a person is liable in respect of a tort committed by his or her employee in the course of his or her employment, and shall, in respect of such a tort, be treated for all purposes as a joint tortfeasor with the member or the protective service officer.
(2) In a claim by the Commonwealth for damages in respect of a tort, an act or omission of a member or a protective service officer in the performance or purported performance of his or her duties as a member or a protective service officer may be relied on as constituting contributory negligence by the Commonwealth if the act or omission could have been so relied on if it had been done by an employee of the Commonwealth in the course of his or her employment.
(3) The liability of the Commonwealth under subsection (1) does not extend to a liability to pay damages in the nature of punitive damages.
Qld
(1) The Crown is liable for a tort committed by any police officer acting, or purporting to act, in the execution of duty as a police officer under this Act in like manner as an employer is liable for tort committed by the employer’s servant in the course of employment.
(2) The Crown is to be treated for all purposes as a joint tortfeasor with the police officer who committed the tort.
(3) In no case does the Crown’s liability for a tort committed by any police officer extend to a liability to pay damages in the nature of punitive damages.
WA
(6) The Crown’s liability under subsection (5) does not extend to exemplary or punitive damages.

* BA, LLB (UWA), BCL, MPhil, DPhil (Oxon); Senior Lecturer in Law, University of Western Australia. My thanks to Daniel Wigston for assistance with the statistical analysis, Professor James Goudkamp for helpful guidance, and the anonymous referees for their insightful comments.

[1] Whitfeld v De Lauret & Co Ltd [1920] HCA 75; (1920) 29 CLR 71, 81 (Isaacs J) (‘Whitfeld’); Uren v John Fairfax & Sons Pty Ltd [1966] HCA 40; (1966) 117 CLR 118, 130 (Taylor J), 149 (Windeyer J) (‘Uren’); Rookes v Barnard [1964] UKHL 1; [1964] AC 1129, 1221 (Lord Devlin) (‘Rookes’).

[2] For detailed discussion of these and other objections to exemplary damages, see, eg, Michael Tilbury and Harold Luntz, ‘Punitive Damages in Australian Law’ (1995) 17(4) Loyola of Los Angeles International and Comparative Law Journal 769, 786–92; Andrew Burrows, ‘Reforming Exemplary Damages: Expansion or Abolition?’ in Peter Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press, 1996) 153; Nicholas J McBride, ‘Punitive Damages’ in Peter Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press, 1996) 175; Rachael Mulheron, ‘Exemplary Damages and Tort: An International Comparison’ [2000] UNDAULawRw 3; (2000) 2 University of Notre Dame Australia Law Review 17; Ernest J Weinrib, ‘Punishment and Disgorgement as Contract Remedies’ (2003) 78(1) Chicago-Kent Law Review 55, 84–102; Allan Beever, ‘The Structure of Aggravated and Exemplary Damages’ (2003) 23(1) Oxford Journal of Legal Studies 87, 94–110; Ralph Cunnington, ‘Should Punitive Damages Be Part of the Judicial Arsenal in Contract Cases?’ (2006) 26(3) Legal Studies 369, 380–4; Robert Stevens, Torts and Rights (Oxford University Press, 2007) 85–8; James Edelman, ‘In Defence of Exemplary Damages’ in Charles EF Rickett (ed), Justifying Private Law Remedies (Hart Publishing, 2008) 225.

[3] See, eg, Gray v Motor Accident Commission [1998] HCA 70; (1998) 196 CLR 1, 6 [12], 9 [20] (Gleeson CJ, McHugh, Gummow and Hayne JJ) (‘Gray’).

[4] See, eg, XL Petroleum (NSW) Pty Ltd v Caltex Oil (Australia) Pty Ltd [1985] HCA 12; (1985) 155 CLR 448, 463 (Gibbs CJ) (‘XL Petroleum’).

[5] The key legislative provisions are summarised below in Part III.

[6] James Goudkamp and Eleni Katsampouka, ‘An Empirical Study of Punitive Damages’ (2018) 38(1) Oxford Journal of Legal Studies 90, 91 n 4. More specifically, this was a study of punitive damages in England, Wales and Northern Ireland. The UK Study omitted Scotland because exemplary damages are unknown in Scots law: at 90.

[7] With the major exception of the USA, and one ‘modest and dated study in Canada’: ibid 91, citing Neil Vidmar and Bruce Feldthusen, ‘Exemplary Damages Claims in Ontario: An Empirical Profile’ (1990) 16(3) Canadian Business Law Journal 262.

[8] There has been no report of an Australian law reform body concerned primarily with exemplary damages. The remedy has been considered briefly, and incidentally, in a number of law reform commission reports primarily concerned with other topics: see, eg, New South Wales Law Reform Commission, Defamation (Report No 11, 1971) [42]–[57]; Law Reform Commission of Tasmania, Compensation for Victims of Motor Vehicle Accidents (Report No 52, 27 October 1987) 45–6; Law Reform Commission, Product Liability (Report No 51, 1 June 1989) 104–6 [8.10]–[8.14]; Law Reform Commission of Victoria, Product Liability (Report No 27, 1 June 1989) 104–6 [8.10]–[8.14].

[9] See, eg, Mulheron (n 2) 51; Beever (n 2) 104; Andrew Phang and Pey-Woan Lee, ‘Restitutionary and Exemplary Damages Revisited’ (2003) 19(1) Journal of Contract Law 1, 36–7.

[10] See below Part IV.

[11] See, eg, Justice James Edelman, McGregor on Damages (Sweet & Maxwell, 20th ed, 2018) 415–18 [13-002]–[13-008]; Cunnington (n 2) 370–3; Alison Doecke, ‘Exemplary Damages: Retribution and Condemnation — the Purpose Controlling the Scope of the Exemplary Damages Award’ [2017] AdelLawRw 4; (2017) 38(1) Adelaide Law Review 87, 88–91; Edelman, ‘In Defence of Exemplary Damages’ (n 2) 227–9; BW Collis, ‘“Tort and Punishment” — Exemplary Damages: the Australian Experience’ (1996) 70(1) Australian Law Journal 47, 47–8.

[12] Whitfeld (n 1).

[13] Ibid 77 (Knox CJ).

[14] Uren (n 1).

[15] Rookes (n 1).

[16] Uren (n 1) 122–3 (McTiernan J), 139 (Taylor J), 146–7 (Menzies J), 148 (Windeyer J), 161 (Owen J). The decision of the High Court was affirmed on appeal by the Privy Council: Australian Consolidated Press Ltd v Uren [1967] UKPCHCA 2; (1967) 117 CLR 221.

[17] Rookes (n 1) 1226–7 (Lord Devlin).

[18] Ibid 1227–8.

[19] See Gray (n 3) 7 [14], 8 [18] (Gleeson CJ, McHugh, Gummow and Hayne JJ), 27 [84] (Kirby J).

[20] AB v South West Water Services Ltd [1993] QB 507, 521–3 (Stuart-Smith LJ), 529–31 (Bingham MR) (‘AB’). The decision in AB was overruled in Kuddus v Chief Constable of Leicestershire Constabulary [2001] UKHL 29; [2002] 2 AC 122, 133–4 [21] (Lord Slynn), 138–40 [38]–[45] (Lord Mackay), 145 [68] (Lord Nicholls), 153 [89] (Lord Hutton), 157–9 [111]–[122] (Lord Scott) (‘Kuddus’).

[21] AB (n 20) 516–17 (Stuart-Smith LJ).

[22] Tilbury and Luntz (n 2) 769.

[23] Gray (n 3) 7 [14] (Gleeson CJ, McHugh, Gummow and Hayne JJ).

[24] Ibid 9–10 [22].

[25] Butler v Fairclough [1917] HCA 9; (1917) 23 CLR 78, 89 (Griffith CJ), quoted in ibid 6 [13]. There may, however, be some room for argument both as a matter of precedent and in principle: see, eg, Katy Barnett and Sirko Harder, Remedies in Australian Private Law (Cambridge University Press, 2nd ed, 2018) 399–402 [14.35]–[14.40].

[26] Principally, breach of fiduciary duty and the equitable duty of confidence: Harris v Digital Pulse Pty Ltd [2003] NSWCA 10; (2003) 56 NSWLR 298 (‘Harris (Appeal)’). However, this issue has yet to be decided by the High Court.

[27] See Lamb v Cotogno [1987] HCA 47; (1987) 164 CLR 1, 9 (Mason CJ, Brennan, Deane, Dawson and Gaudron JJ).

[28] See, eg, Zorom Enterprises Pty Ltd v Zabow (2007) 71 NSWLR 354, 366–7 [44]–[46] (Basten JA).

[29] Ibid.

[30] Gray (n 3) 14 [40]–[43] (Gleeson CJ, McHugh, Gummow and Hayne JJ), 31–4 [92]–[98] (Kirby J), 50–1 [143] (Callinan J).

[31] Fontin v Katapodis [1962] HCA 63; (1962) 108 CLR 177, 186–7 (Owen J).

[32] XL Petroleum (n 4) 472 (Brennan J); Pollack v Volpato [1973] 1 NSWLR 653, 657–8 (Hutley JA) (‘Pollack’).

[33] XL Petroleum (n 4) 471 (Brennan J).

[34] Ibid 463 (Gibbs CJ).

[35] The annexed table is not intended to be exhaustive.

[36] Law Reform (Miscellaneous Provisions) Act 1934, 24 & 25 Geo 5, c 41, s 1(2)(a).

[37] Administration and Probate Act 1935 (Tas) s 27(3)(a).

[38] Law Reform (Miscellaneous Provisions) Act 1955 (ACT) s 5(a); Civil Law (Wrongs) Act 2002 (ACT) s 16(2) (‘ACT Wrongs Act’); Law Reform (Miscellaneous Provisions) Act 1944 (NSW) s 2(2)(a)(i); Law Reform (Miscellaneous Provisions) Act 1956 (NT) s 6(1)(a); Succession Act 1981 (Qld) s 66(2)(b); Administration and Probate Act 1958 (Vic) s 29(2)(a); Law Reform (Miscellaneous Provisions) Act 1941 (WA) s 4(2)(a).

[39] This can be compared with other jurisdictions: see, eg, Ontario Law Reform Commission, Exemplary Damages (Report, 1 June 1991); Law Commission for England and Wales, Aggravated, Exemplary and Restitutionary Damages (Report No 247, 16 November 1997); Law Reform Commission of Ireland, Aggravated, Exemplary and Restitutionary Damages (Report No 60-2000, May 2000).

[40] David Rolph, ‘Australia’s Defamation Laws Are Ripe for Overhaul’, The Sydney Morning Herald (online, 9 December 2018) <https://www.smh.com.au/national/australia-s-defamation-laws-are-ripe-for-overhaul-20181207-p50kwk.html>, archived at <https://perma. cc/7F8J-RCWE>.

[41] ACT Wrongs Act (n 38) s 139H; Defamation Act 2005 (NSW) s 37 (‘NSW Defamation Act’); Defamation Act 2006 (NT) s 35; Defamation Act 2005 (Qld) s 37; Defamation Act 2005 (SA) s 35; Defamation Act 2005 (Tas) s 37; Defamation Act 2005 (Vic) s 37; Defamation Act 2005 (WA) s 37.

[42] Review of the Law of Negligence (Final Report, September 2002) 226 [13.165].

[43] Ibid 225–6 [13.164].

[44] Ibid 226 [13.165].

[45] Civil Liability Act 2002 (NSW) s 21; Personal Injuries (Liabilities and Damages) Act 2003 (NT) s 19; Civil Liability Act 2003 (Qld) s 52(1).

[46] Trade Practices Amendment (Personal Injuries and Death) Act (No 2) 2004 (Cth) sch 1 item 9, inserting Trade Practices Act 1974 (Cth) s 87ZB. The same provision is now found in Competition and Consumer Act 2010 (Cth) s 87ZB (‘CCA’).

[47] And also postdated.

[48] Road Transport (Third-Party Insurance) Act 2008 (ACT) s 22(1)(e); Motor Accidents Act 1988 (NSW) s 81A; Motor Accidents Compensation Act 1999 (NSW) s 144; Motor Accident Injuries Act 2017 (NSW) s 4.20; Motor Accidents (Compensation) Act 1979 (NT) s 6(3)(a); Motor Accident Insurance Act 1994 (Qld) sch s 3(3) (‘Qld Motor Accident Act’); Motor Vehicles Act 1959 (SA) s 113A; Moto r Accidents (Liabilities and Compensation) Act 1973 (Tas) s 14(6); Transport Accident Act 1986 (Vic) s 93.

[49] See, eg, Qld Motor Accident Act (n 48) sch s 3(3).

[50] And also postdated.

[51] Workers Compensation Act 1987 (NSW) s 151R; WorkCover Queensland Act 1996 (Qld) s 319; Workers’ Compensation and Rehabilitation Act 2003 (Qld) s 306B (‘Qld Workers’ Act’); Return to Work Act 2014 (SA) s 81; Accident Compensation Act 1985 (Vic) s 134AB(22).

[52] See, eg, Qld Workers’ Act (n 51) s 306B.

[53] Stevens (n 2) 88.

[54] See, eg, Barnett and Harder (n 25) 397 [14.32], citing Harold Luntz et al, Torts: Cases and Commentary (LexisNexis Butterworths, 8th ed, 2017) 519 [8.1.11].

[55] See below Part V(C).

[56] Australian Federal Police Act 1979 (Cth) s 64B(3); Weapons Act 1990 (Qld) s 165(3); Police Act 1892 (WA) s 137(6).

[57] Goudkamp and Katsampouka (n 6) 96.

[58] The UK Study also excluded tribunal decisions on the ground that tribunals are fundamentally different creatures from courts, which might diverge in terms of how they apply the law relating to exemplary damages: ibid. In Australia, there are few tribunals with jurisdiction to award exemplary damages: see below Part VIII.

[59] Via Westlaw AU and Lexis Advance.

[60] The UK Study did not include appeals as ‘cases that are the subject of an appeal may sometimes be atypical’: Goudkamp and Katsampouka (n 6) 96.

[61] The same search terms were used in the UK Study: ibid 97.

[62] Cf ibid, which yielded about 650 cases.

[63] By the author personally, with some support from a research assistant.

[64] Therefore excluding, for example, interlocutory decisions.

[65] For example, ‘additional damages’ under intellectual property statutes: see, eg, Trade Marks Act 1995 (Cth) s 126(2).

[66] See above Part III.

[67] See, eg, NSW Defamation Act (n 41) s 37 (and other state and territory equivalents), which excludes exemplary damages in defamation claims.

[68] Harris (Appeal) (n 26) 312 [57]–[61] (Spigelman CJ).

[69] As the authors of the UK Study note, in such cases, the court may be less careful in considering exemplary damages, distorting the sample: Goudkamp and Katsampouka (n 6) 97.

[70] Cf ibid, in which there were 100 cases: at 98.

[71] Cf ibid, in which appeals were not included and some first instance decisions were accessed indirectly through appellate decisions: at 99.

[72] The same exceptions were made in the UK Study: ibid 98.

[73] Ibid.

[74] Ibid.

[75] Riches v News Group Newspapers Ltd [1985] EWCA Civ 20; [1986] QB 256, 289 (Parker LJ) (‘Riches’).

[76] Goudkamp and Katsampouka (n 6) 99.

[77] This is not uncommon in Australia: see below n 171 and accompanying text.

[78] This is also not uncommon in Australia: see below n 166 and accompanying text.

[79] Also not uncommon in Australia: see below n 172 and accompanying text.

[80] See below nn 166–7, 171–3 and accompanying text.

[81] Goudkamp and Katsampouka (n 6) 98.

[82] See generally ibid 102–9.

[83] Ibid 99.

[84] See below Part V(B).

[85] ‘Inflation Calculator’, Reserve Bank of Australia (Web Page) <https://www.rba.gov.au/calculator/>, archived at <https://perma.cc/JT8R-FGMP>.

[86] The UK Study reported compensatory damages including any aggravated damages in this column: Goudkamp and Katsampouka (n 6) 100–1.

[87] Uren (n 1) 149 (Windeyer J).

[88] Ibid.

[89] See below nn 157–62 and accompanying text.

[90] This was in part a function of appeals being included in this study. Also, relatively few civil trials are decided by jury in Australia today: see, eg, Barnett and Harder (n 25) 394–5 [14.22].

[91] Goudkamp and Katsampouka (n 6) 101.

[92] Ibid.

[93] Ibid 102.

[94] The statistician used IBM SPSS Statistics version 25.

[95] For an explanation of the Pearson’s chi-squared test, see generally Lee Epstein and Andrew D Martin, An Introduction to Empirical Legal Research (Oxford University Press, 2014) 168–70.

[96] For an explanation of analysis of variance (‘ANOVA’) tests, see generally Morris H DeGroot and Mark J Schervish, Probability and Statistics (Addison-Wesley, 4th ed, 2012) 754–63.

[97] For an explanation of linear regression analysis, see generally Epstein and Martin (n 95) ch 8.

[98] This was an extreme outlier claim, a huge award in the Federal Court in a passing off case, which plainly affected the mean award: Deckers Outdoor Corporation Inc v Farley [No 5] [2009] FCA 1298; (2009) 262 ALR 53, 78 [115] (Tracey J) (‘Deckers’).

[99] One of four equal parts into which a population of data can be divided.

[100] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of 0.088.

[101] This was plainly affected by the extreme outlier claim mentioned earlier: see above n 98.

[102] An ANOVA test yielded a P value of 0.429 (where F(7, 112) = 1.008).

[103] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of less than 0.001.

[104] An ANOVA test yielded a P value of 0.758 (where F(8,11) = 0.622).

[105] See above Part III.

[106] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of 0.005.

[107] An ANOVA test generated a P value of 0.021 (where F(2,117) = 4.004).

[108] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of less than 0.001.

[109] A t test yielded a P value of 0.530 (where t = 0.631).

[110] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of 0.001.

[111] Linear regression analysis yielded a P value of 0.731 (where beta coefficient = −0.27, R2 = 0.001% and adjusted R2 = −0.006%).

[112] See above nn 86–9 and accompanying text.

[113] A Pearson’s chi-squared test with Yates’s continuity correction yielded a P value of 0.063.

[114] Linear regression analysis yielded a P value of less than 0.001 (where beta coefficient = 0.400, R2 = 0.160% and adjusted R2 = 0.156%).

[115] See, eg, Goudkamp and Katsampouka (n 6) 91 n 4.

[116] Ibid 110.

[117] See above Part V(A)(1).

[118] Likely to involve an award of (or the refusal to award) compensatory damages.

[119] Supreme Court of New South Wales, 2016 Annual Review (Report, 2016) 46.

[120] See, eg, Backwell v AAA [1997] 1 VR 182, 205–6 (Ormiston JA).

[121] Engle v RJ Reynolds Tobacco Co, 122 F Supp 2d 1355 (SD Fla, 2000).

[122] Deckers (n 98). See at

[123] See above Part V(A)(1).

[124] Goudkamp and Katsampouka (n 6) 114.

[125] Australian Bureau of Statistics, Regional Wage and Salary Earner Statistics, Australia: Electronic Publication, 2000–01 (Catalogue No 5673.0.55.001, 19 December 2003).

[126] Australian Bureau of Statistics, Average Weekly Earnings, Australia, Nov 2016 (Catalogue No 6302.0, 23 February 2017). The annual figure was calculated by multiplying the average weekly total earnings by 52, and adjusting for inflation to 2017.

[127] See below Part VI(B).

[128] Goudkamp and Katsampouka (n 6) 115, citing Theodore Eisenberg and Michael Heise, ‘Judge–Jury Difference in Punitive Damages Awards: Who Listens to the Supreme Court?’ (2011) 8(2) Journal of Empirical Legal Studies 325, 331.

[129] See Gray (n 3) 5–12 [8]–[31] (Gleeson CJ, McHugh, Gummow and Hayne JJ).

[130] New South Wales, Northern Territory, Queensland, South Australia and Western Australia.

[131] Northern Territory, Queensland and Western Australia.

[132] Principally breach of contract and equitable wrongdoing: Harris (Appeal) (n 26) 312 [57]–[61] (Spigelman CJ).

[133] AB (n 20). This ‘cause of action’ test was overruled in Kuddus (n 20).

[134] Whitfeld (n 1) 77 (Knox CJ).

[135] There was no such statistically significant association between the cause of action category and the amount of exemplary damages awarded.

[136] See above Part III.

[137] Gunston v Davies Brothers Ltd (2012) 21 Tas R 256 (‘Gunston’) (arising out of publications in 2002 and 2003).

[138] Gacic v John Fairfax Publications Pty Ltd [2015] NSWCA 99; (2015) 89 NSWLR 538 (‘Gacic’) (also arising out of a publication in 2003).

[139] See, eg, MacDonald v Public Trustee [2010] NSWSC 684 (in respect of conduct dating back 40 years); Nationwide News Pty Ltd v Naidu [2007] NSWCA 377; (2007) 71 NSWLR 471 (in respect of conduct which took place in the 1990s).

[140] See above n 106 and accompanying text.

[141] See above Part V(C).

[142] See below Part VI(A)(5).

[143] See above n 107 and accompanying text.

[144] See above Part V(C).

[145] See XL Petroleum (n 4) 472 (Brennan J).

[146] ‘If exemplary damages are to fulfil their threefold purpose, they must not merely irritate, they must sting’: Digital Pulse Pty Ltd v Harris (2002) 166 FLR 421, 440 [133] (Palmer J) (Supreme Court of New South Wales) (‘Harris (Trial)’). See also ibid; Pollack (n 32) 657–8 (Hutley JA).

[147] Goudkamp and Katsampouka (n 6) 112–13.

[148] See above Part V(D).

[149] See, eg, Adams v Kennedy [2000] NSWCA 152; (2000) 49 NSWLR 78 (‘Adams’); Cunningham v Traynor [2016] WADC 168 (‘Cunningham’); Majindi v Northern Territory [2012] NTSC 25; (2012) 31 NTLR 150 (‘Majindi’).

[150] See above Part V(D).

[151] See above n 56 and accompanying text.

[152] See above Part V(D).

[153] XL Petroleum (n 4) 471 (Brennan J); Harris (Appeal) (n 26) 345 [254] (Heydon JA), quoting Harris (Trial) (n 146) 440 [133] (Palmer J).

[154] Rookes (n 1) 1228 (Lord Devlin).

[155] Ibid.

[156] See, eg, Gacic (n 138); Majindi (n 149); New South Wales v Quirk [2012] NSWCA 216 (‘Quirk’); Schmidt v Argent [2003] QCA 507 (‘Schmidt’); Van der Poel v Hall [2009] NSWDC 50 (‘Van der Poel’).

[157] These claims were omitted for the purposes of some of the statistical analysis, as it was not possible in these claims to ascertain the amount of exemplary damages awarded.

[158] [2018] NSWCA 320 (‘Cuthbertson’).

[159] [2006] HCA 57; (2006) 229 CLR 638, 648 [35] (Gleeson CJ, Gummow, Kirby, Heydon and Crennan JJ) (‘Ibbett’).

[160] Cuthbertson (n 158) [115]–[116] (McColl JA).

[161] Cf Broome v Cassell & Co Ltd [1972] UKHL 3; [1972] AC 1027, 1082 (Lord Hailsham LC), 1099 (Lord Morris) (‘Broome’); Riches (n 75) 268 (Stephenson LJ). These UK cases provide conflicting authority on whether exemplary and compensatory damages should be awarded as a single sum or as separate awards. See also Edelman, McGregor on Damages (n 11) 1848–9 [52-047].

[162] In these claims, for the purposes of some of the statistical analysis, compensatory damages were treated as aggravated damages, as an award of aggravated damages is compensatory but compensatory damages are not aggravated damages: see above nn 86–9 and accompanying text.

[163] Broome (n 161) 1062–3 (Lord Hailsham LC). See also Edelman, McGregor on Damages (n 11) 440–1 [13-043]; Stevens (n 2) 87; Burrows (n 2) 172. See Mulheron (n 2) 46–7.

[164] Burrows (n 2) 172.

[165] Stevens (n 2) 87; Mulheron (n 2) 46–7.

[166] Commonwealth: see, eg, Deckers (n 98) 78 [115], 81 [140] (Tracey J). New South Wales: see, eg, James v Hill [2004] NSWCA 301, [45]–[47] (Tobias JA). Victoria: see, eg, De Reus v Gray [2003] VSCA 84; (2003) 9 VR 432, 457 [39] (Winneke P). Queensland: see, eg, Schmidt (n 156) [51] (Duntley J). Tasmania: see, eg, Gunston (n 137) 291–2 [204]–[206] (Crawford CJ). Northern Territory: see, eg, Majindi (n 149) 173 [74] (Mildren J).

[167] See Law Commission for England and Wales (n 39) 80 [1.164].

[168] Riches (n 75) 289 (Parker LJ). See Burrows (n 2) 172–3; Stevens (n 2) 87–8.

[169] Burrows (n 2) 172.

[170] Ibid; Mulheron (n 2) 47.

[171] New South Wales: see, eg, Van der Poel (n 156) [122]–[130] (Sidis DCJ). Victoria: see, eg, Victoria v Horvath [2002] VSCA 177; (2002) 6 VR 326, 339–41 [29]–[36] (Winneke P, Chernov and Vincent JJA) (‘Horvath’). Western Australia: see, eg, Cunningham (n 149) [919]–[921], [1080]–[1088] (Davis DCJ). South Australia: see, eg, White v South Australia [2010] SASC 95; (2010) 106 SASR 521, 597–602 [471]–[526] (Anderson J).

[172] Commonwealth: see, eg, Ibbett (n 159) 644 [20]–[21] (Gleeson CJ, Gummow, Kirby, Heydon and Crennan JJ). New South Wales: see, eg, Quirk (n 156) [148]–[163] (Tobias AJA). Victoria: see, eg, Horvath (n 171) 339–40 [29] (Winneke P, Chernov and Vincent JJA).

[173] Adams (n 149) 87 [36] (Priestley JA).

[174] The period of this study is one year longer, 2000–16, rather than the 2000–15 period of the UK Study: see Goudkamp and Katsampouka (n 6) 96.

[175] Ibid 98.

[176] Ibid 103.

[177] Using currency converter on XE Converter at 25 January 2019: ‘XE Currency Converter’, XE (Web Page) <https://xe.com>.

[178] Goudkamp and Katsampouka (n 6) 103.

[179] Phang and Lee (n 9) 18.

[180] See above Part V(B).

[181] Goudkamp and Katsampouka (n 6) 104–5.

[182] Ibid 104.

[183] Ibid 105.

[184] Ibid 105, 113–14.

[185] CCA (n 46) sch 2.

[186] See above Part V(C).

[187] Goudkamp and Katsampouka (n 6) 106.

[188] Rookes (n 1) 1226 (Lord Devlin).

[189] Goudkamp and Katsampouka (n 6) 106–7.

[190] Ibid 116.

[191] See above Part V(E).

[192] Goudkamp and Katsampouka (n 6) 108–9.

[193] Ibid 109.

[194] See ibid.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2020/8.html