Melbourne University Law Review
[For 30 years Harold Luntz has been a major participant in personal injury law reform debates and a passionate advocate of the abolition of tort law and its replacement by a no-fault compensation scheme. The aim of this article is to place the 2002 Review of the Law of Negligence in the wider context of those debates and to assess the prospects for more radical reform of personal injury law in Australia. It also addresses the complex issue of the relationship between tort liability and the cost of liability insurance in the context of the so-called insurance ‘crisis’.]
Harold Luntz’s tort scholarship is characterised by admirable qualities of its author, including an encyclopaedic knowledge of the law, meticulous attention to detail, utter respect for truth and, perhaps above all, a passionate advocacy of no-fault personal injury compensation schemes. Harold’s views about ‘reform’ of personal injury law deserve respect precisely because they are based on an understanding of the present system that is second to none. However, as Harold has recently written (with typical understatement), the replacement of tort law with a no-fault system in Australia ‘is unlikely in the present political climate’; and so for him (and for those of us who share his vision), the real question concerns the best way of promoting the ideal within existing political constraints.
Against this background, the aim of this article is to assess the place of the recent Review of the Law of Negligence in the context of debates about compensation for personal injury that have taken place over the past 40 years in the common law world. My main argument is that, in order to gain a sound understanding of the Review, careful attention must be paid to the political environment in which it was conducted. I also suggest that, although the general ideology of the Review appears, on the surface at least, to be diametrically opposed to that which motivates proposals for no-fault compensation schemes, there is still hope that the force of the well-known and empirically supported arguments against the tort system may yet bear fruit. The fact that there has been so much public debate about the tort system in the past couple of years (albeit of highly variable quality) shows that the issues involved can have great political salience. It is when strong political will and sound policy-making come together that real legal progress can be made.
Before launching into the main discussion, Part II seeks to destabilise the concept of law reform and to replace it with a more complex picture of ongoing debates about the terms of social life and, in particular, about the way in which the costs of illness and disability are dealt with. Part III contains a brief history of modern thinking and policy-making in the area of personal injury law. Part IV examines the genesis of the Review and discusses the central issue of the relationship between tort law and liability insurance. Part V explains certain features of the Review in the context of the political environment at the time. In Part VI, I suggest a strategy for exploiting current dissatisfaction with tort law that offers some hope of advancing the cause of no-fault compensation.
‘Reform’ denotes change and connotes improvement. In retrospect, of course, changes may be seen not to have improved matters and, in prospect, people may disagree about what improvements, if any, are needed and how to achieve them. People may also disagree about what counts as an improvement. According to the ‘Pareto’ criterion espoused by some economists, a change will constitute an improvement only if it makes at least one person better off and no-one worse off. An alternative to this extremely demanding test is the ‘Kaldor-Hicks’ criterion, which requires only that the aggregate benefits of a change outweigh its costs — in other words, that the change produce more winners than losers. Popular with lawyers are what I have elsewhere called ‘single-factor consequentialist’ arguments. For instance, in contemporary Australian debates, a key expectation nurtured by proponents of changes to tort law is that reform will lead to a lowering of liability insurance premiums. Many lawyers are also attracted by non-consequentialist arguments for change based on ideas such as ‘coherence’ and ‘consistency’ in the law. Underlying such ideas is the ‘mothers-and-apple-pie’ concept of formal justice, which requires treating like cases alike and unlike cases differently. The devil, of course, resides in the detailed elaboration of what constitutes likeness and unlikeness. Ultimately, all of these approaches to legal change rest on value judgments about how the benefits and burdens of social life (including legal rights and obligations) should be distributed.
The basic point is that ‘reform’ is a contested concept, as recent Australian debates about personal injury law make clear. ‘Policy entrepreneurs’ participating in these debates fall into three main groups that we might loosely call conservatives, radicals and moderates.
There are two species of conservatives — compensationists and economic rationalists. Compensationists oppose legislative change to personal injury law on the basis that ‘what ain’t broke don’t need fixing’. For them, the main purpose of personal injury law is to compensate people injured by the wrongdoing of others. To this end, tort law has been developed and refined by courts in the course of the past century or so with minimal parliamentary intervention. Non-judicial change is likely to upset the fine balance struck by tort law between the interests of injurers and the injured, and so should be opposed. The leading representative of this species of conservative in Australia is the Australian Plaintiff Lawyers Association (‘APLA’).
In the opinion of the economic rationalists, the prime function of personal injury law is risk-management, not compensation. They oppose reducing the scope of tort liability or the quantum of damages, arguing that this will reduce the efficacy of personal injury law as a regulatory tool and expose ‘consumers’ to unacceptable threats to their personal health and safety. The chief spokesperson for the economic rationalists in recent Australian debates has been the Australian Competition and Consumer Commission (‘ACCC’).
Secondly, there are the radicals. They believe that tort law is about as bad as it could be, whichever way you look at it. As a compensation system it is inefficient and extremely expensive; its efficacy as a regulatory tool is, at best, doubtful; it unfairly discriminates between the sick and injured on the basis of the cause of their disabilities; and it embodies concepts of wrongdoing that bear little relation to ‘moral’ ideas of fault. They favour its replacement — in as many areas as possible — by some form of no-fault scheme of support for the disabled. Harold Luntz is the leading Australian radical, although this general approach is common amongst academics who specialise in personal injury law in Australia. Radicals can be divided into those, such as myself, who prefer a social security-type replacement for the tort system, and those, such as Patrick Atiyah, who prefer a market-based solution.
Finally, we have the moderates, of which there are also two types. One group believes that too much is spent on compensating injured people through tort law, and that something needs to be done about it by reducing the scope of liability and the quantum of damages. The other species sees the trouble with tort law as being that it strikes the wrong balance between injurers and injured people in terms of their respective responsibilities to take care for themselves and for others. Putting the point crudely, this group would say that tort law is too pro-claimant. As we will see later, both of these strands of thinking were reflected in the terms of reference of the Review.
Modern debates about personal injury law date from the 1960s. The major intellectual breakthrough involved a shift from thinking about tort law primarily as a system of rules and principles of personal responsibility for the infliction of harm on others, to thinking about it as one, but only one, mechanism for achieving goals such as compensation and deterrence. Especially influential in this regard was the work of Guido Calabresi who, in a series of articles in the 1960s and then in his classic book, The Costs of Accidents, published in 1970, argued that ‘the time has come for a full reexamination of what we want a system of accident law to accomplish and for an analysis of how different approaches to accidents would accomplish our goals.’ ‘I take it as axiomatic’, he said, ‘that the principal function of accident law is to reduce the sum of the costs of accidents and the costs of avoiding accidents.’
Calabresi identified three categories of accident costs: the (‘primary’) cost of reducing the number and severity of accidents; the (‘secondary’) cost of compensating accident victims; and the (‘tertiary’) administrative costs of reducing the number and severity of accidents and of compensating accident victims. He argued that the overall policy goal in designing a system for dealing with the social problem of ‘accidents’ was to minimise the sum total of these three categories of costs. Calabresi dubbed the cost of reducing the number and severity of accidents ‘primary’ because, in his view, achieving such reduction was the main goal of accident law: prevention is better than cure. This normative starting position inevitably contributed to his generally negative assessment of what he called ‘the fault system’. Calabresi analysed the contribution that the fault system makes to reducing the costs of accidents in terms of two concepts: ‘general’ (or ‘market’) deterrence and ‘specific’ deterrence. According to the general deterrence theory of accident law, the imposition of tort liability creates incentives for the taking of precautions against risks of harm only when they are economically justifiable. According to the specific deterrence theory, tort law operates by announcing standards of conduct and providing a mechanism for enforcing those standards. Calabresi’s conclusion was that accident law performed badly in terms of both theories.
By contrast, Terence Ison’s book, The Forensic Lottery, was subtitled ‘A Critique on Tort Liability as a System of Personal Injury Compensation’ (emphasis added). Ison clearly associated himself with the intellectual breakthrough mentioned earlier by saying that his book was about tort liability ‘not ... as a branch of private law, but ... as one of the media of personal injury compensation.’ However, whereas Calabresi offered a theoretical framework for evaluating accident law, Ison offered a practical proposal for replacing tort law with a no-fault system of compensation for ‘diseases and violent injuries causing disablement and death’. Ison’s view was that because the deterrent value of tort law was ‘negligible’, its abolition would not call for serious consideration of what to do to reduce the number and severity of accidents. For him, deterrence was not an important concern of personal injury law reform.
The fact that Calabresi, writing from an American perspective, placed accident reduction at the top of his agenda while Ison (writing from a Commonwealth perspective) put compensation first may be partly explicable by the fact that, since the late 19th century, regulation of social activity had been the primary form of governmental intervention in the economy in the United States. By contrast, public ownership of means of production, and government provision of social security and welfare services, characterised the British economy, especially in the period after the Second World War. United States tort scholarship is generally much more oriented towards viewing tort law as a tool for regulating individual and, especially, corporate behaviour, than Commonwealth tort scholarship, most of which is more inclined to treat tort law as primarily concerned with allocating what Calabresi called ‘secondary’ accident costs.
Undoubtedly the most influential exposition of a ‘functional’ approach to personal injury law was Patrick Atiyah’s Accidents, Compensation and the Law (1970). As its title suggests, Atiyah’s contribution (like Ison’s) was more concerned with secondary than primary accident costs, and with meeting such costs as opposed to reducing them. Indeed, Atiyah discussed Calabresi’s general deterrence theory at length, arguing that ‘one of our principal problems is going to be that of balancing the requirements of general deterrence against those of loss distribution’, the latter being that secondary accident costs ‘should be spread over as wide a segment of the population as possible.’ Atiyah made both ‘internal’ and ‘external’ criticisms of tort law as a compensation system. The internal critique was that tort liability was primarily based on a morally objectionable concept of ‘fault’. The external critique was twofold. First, Atiyah argued, the administrative costs of the tort system are extremely high relative to the amounts of compensation paid out and to the administrative costs of providing analogous social security benefits to the sick and disabled. Second, he said, the tort system provides very generous compensation to a relatively very small group of injured and disabled people. Not only are most disabled people unable to meet the criteria for obtaining tort compensation, but also only a relatively small proportion of those who could, in theory, meet the criteria actually recover tort compensation because the ‘tort system’ presents many barriers to successful enforcement of the rights conferred by tort law.
Concurrent with this ferment in the academy were important developments in the world outside. Serious dissatisfaction with the worker’s compensation system in New Zealand in the 1960s led to the appointment of a Royal Commission chaired by Sir Owen Woodhouse, and the eventual enactment of the Accident Compensation Act 1972 (NZ). The Act provided for a no-fault accident compensation scheme, which came into operation in 1974 and remains the most comprehensive scheme of its kind operating anywhere in the world. A notable limitation of the New Zealand scheme is that it deals mainly with injury by accident; its coverage of diseases is limited to occupational conditions. In the early 1970s the Australian Whitlam Labor government appointed a committee, chaired by Woodhouse, to develop proposals for a no-fault compensation scheme covering diseases as well as accidents. However, this initiative was a casualty of the 1975 constitutional crisis and the subsequent election of a Liberal government, and has never been revived.
In Great Britain, the catalyst for action was the difficulty faced by congenitally-disabled children in using the tort system against manufacturers of the drug thalidomide. A Royal Commission, chaired by Lord Pearson, reported in 1978. Its main importance lay, and continues to lie, not in its recommendations — which were piecemeal and bore very little fruit — but in the extensive empirical studies that it sponsored, which strongly supported the broad thrust of Ison’s and Atiyah’s critiques of the tort system.
The 1970s and early 1980s saw the establishment, worldwide, of many no-fault compensation schemes, most commonly limited to automobile or medical accidents. Some such schemes entirely replace the tort system in the area of their operation (for example, the motor accident scheme in the Northern Territory), whereas others supplement it, typically by providing a basic set of benefits for all injured persons while allowing the more seriously injured recourse to the tort system to supplement the no-fault benefits (an example being the transport accident scheme in Victoria). But by the late 1980s, the move to no-fault had run out of steam. Proposals made by the New South Wales Law Reform Commission in 1984 for a no-fault transport accident scheme to replace the tort system proved abortive, and by the 1990s the political climate had become distinctly hostile to the sort of communitarian solutions to the ‘problem of accidents’ that had been favoured by most reform-minded lawyers in Great Britain and the old Commonwealth since the 1960s.
Nevertheless, the defects of the tort system, which had been understood and increasingly well documented for about 30 years, remained, as did the policy questions about how to deal with them. The time was ripe for a radical reconsideration of the problem of accidents. Perhaps not suprisingly, it was Patrick Atiyah, lured out of retirement in 1996 for a guest appearance at a seminar held at All Souls’ College in Oxford, who set the process in motion, first in the proceedings of that seminar, and then in The Damages Lottery. The echo of Ison’s 1967 book in the title of Atiyah’s signals that, in many respects, his critique of the tort system in the late 1990s was the same as that first elaborated by him in 1970 in Accidents, Compensation and the Law. But whereas 30 years earlier the solution favoured by Atiyah, reflecting the temper of the times, involved an extension of the welfare state along the lines of the New Zealand accident compensation scheme, in 1997 — once again, reflecting the contemporary, and very different, political environment — the focus of his proposals was on individual self-protection and harnessing the resources and potential of the commercial insurance market.
More importantly, perhaps, Atiyah developed a new critique of the tort system encapsulated in the idea of ‘the blame culture’. In recent years, he argued, tort principles of liability and assessment of damages had been increasingly ‘stretched’ by the courts, typically in favour of accident victims and to the detriment of defendants, such as doctors and public authorities. ‘[I]ndividuals’, he said, ‘must accept responsibility for their own problems.’ Some judges, he thought, were ‘still too much under the influence of the vaguely left-wing welfare culture of the 1960s, even though ... the public itself [has] ... now moved away from it.’ Like Ison in 1967, Atiyah thought it doubtful that tort law played any useful role in reducing the number of accidents and that ‘little would be lost by getting rid of it.’
The fertile ground needed for these ideas to bear fruit in Australia was laid by a series of events in 2001–02, including the attack on the World Trade Centre and consequent instability in financial markets, especially in the United States; the collapse of the HIH insurance company (which held more than 20 per cent of the Australian public liability insurance market) and of United Medical Protection Limited (‘UMP’), the country’s largest provider of medical indemnity; and sudden, large increases in the cost of some lines of liability insurance, especially public liability and medical indemnity cover. Even though personal injury law is predominantly a state responsibility, the political pressure generated by these events was so great and so widespread that the federal government took the lead early in 2002 in developing a nationally coordinated response to what was seen as a pressing issue of great importance by influential groups, such as the medical profession and the adventure tourism industry, and which was depicted as posing a threat to the very fabric of society, especially in small communities of rural and regional Australia. For a period in 2002, the ‘insurance crisis’ was one of the most important and widely discussed issues in Australian domestic politics.
Central to debates about the insurance crisis were assertions of a positive correlation between the rules and principles of tort law and their application in the tort system, and the sudden blowout in the cost of liability insurance. The argument was that during the 1990s, changes in the rules of tort liability and in judicial attitudes had made it easier for claims to succeed, and that changes in the rules of assessment of damages and in judicial attitudes had resulted in ever-increasing levels of compensation. These developments, it was suggested, had made a significant — if not the most significant — contribution to rises in insurance premiums, especially in 2002. Unfortunately, the available data does not enable the validity of this argument to be properly assessed. In this, as in many other respects, perceptions are at least as important as reality. In this regard, it is worth examining a report prepared by Trowbridge Consulting Ltd (an arm of the Deloitte Touche Tohmatsu group) for the Insurance Issues Working Group of the Heads of Treasuries. This report was presented to a meeting of all nine Australian governments on 30 May 2002 and it was on the basis of the recommendations made therein that the Review was established.
The authors of the report concluded, on the basis of an examination of available statistical information about personal injury litigation, that ‘[o]verall ... there has been a steady increase in public liability insurance bodily injury claims over the last five to ten years. There is no evidence of an “explosion of litigation” in recent years.’
The important part of this conclusion is the second sentence — ‘[t]here is no evidence of an “explosion of litigation” in recent years’ — because what needs to be explained is not a steady, incremental increase in liability insurance premiums throughout the 1990s (which, the evidence suggests, did not occur), but the sudden, large increases experienced in 2002 in particular.
In light of this conclusion about trends in legal liability, how did the authors of the Trowbridge Report support the recommendation for a review of the law of negligence? Three ‘arguments’ were offered. First, it was said that ‘the evidence indicates a gradual “shift” or “stretching” of the interpretation of negligence over several decades so that there are cases succeeding today that would not have succeeded at times in the past.’ Secondly, the report argued
that there is a mismatch between the current state of the law of negligence and the expectations of the community, in terms of the balance between cost and that which the community views as legitimate negligence or breach of duty of care, deserving of fault-based compensation against a wrong-doer.
Finally, it was suggested that
tort reform is ... a means of curbing escalation of the underlying cost of claims, thereby increasing the predictability of claims outcomes. Analysis has shown that the cost of public liability claims has been rising for a number of years at a growth rate above the level of inflation.
It is noteworthy that none of these arguments assumes any relationship between liability law and the tort system on the one hand, and volatility in the insurance market (the insurance ‘crisis’) on the other. Given the importance of this issue in fuelling calls for reform of tort law, and the intuitive plausibility of assertions of a direct link between the cost of tort liability and the cost of liability insurance, it is worth considering the matter in more detail. It is important for a proper assessment of the Review.
Liability insurance ‘crises’, characterised by sudden large increases in premiums, are not new. For instance, there have been several in the United States in the last 30 years. In order to understand the nature of such events, it is helpful to distinguish between three different ways of viewing liability insurance: (i) as a risk-spreading device; (ii) as a compensation mechanism; and (iii) as a commodity. These three different perspectives reflect the fact that liability insurance sets up a triangular relationship between the injurer, the injured and the insurer. For the injurer, liability insurance is primarily a risk-spreading device; for the injured it is a compensation mechanism; and for the insurer it is essentially a commodity.
At the time the liability insurance market started to develop in the late 19th century, liability insurance was conceived primarily as a means by which potential injurers could spread the risk of incurring liability. As between the insurer and the insured, liability insurance is much like any other type of insurance — the insured pays a premium in return for indemnity against specified losses. Insurance is a contract and, as in the case of the typical contract, people are basically free to enter (or not to enter) liability insurance contracts as they choose. In theory, if the cost of liability insurance accurately reflects the risk that the insured will incur liability, having insurance should not alter the insured’s incentive to avoid incurring liability. In reality, ‘moral hazard’ is a major problem against which insurers take various precautions. In reality, too, for various reasons the cost of a liability insurance policy often does not accurately reflect the insured’s risk of incurring liability. Some writers have turned the risk-spreading function of liability insurance from a necessity into a virtue by arguing that certain sorts of liability — strict liability, for instance, and liability for very large sums — is only fair to the extent that the injurer is able to cover the risk of that liability by insurance.
As the liability insurance market grew and the toll of death and injury in the workplace and on the roads became an increasingly serious social problem, attention turned to the compensatory function of liability insurance. At various times in various jurisdictions, statutory provisions were enacted requiring vehicle owners and employers to purchase insurance against liability for personal injury resulting from transport and workplace accidents. The advent of compulsory liability insurance — liability insurance as social welfare, we might say — transformed people’s understanding of what personal injury law was for and about. Tort law was no longer seen as a set of rules and principles of personal responsibility for harm caused to others, but rather as a form — albeit a very expensive and convoluted form — of social security. Compulsory liability insurance premiums came to be viewed as a sort of tax and the relationship between tort law and liability insurance was reversed. Whereas liability insurance had originally been seen as a useful but optional adjunct to tort law, by the 1970s Patrick Atiyah could interpret the tort system as an insurance-based administrative compensation scheme with a judicial dispute resolution mechanism attached.
By contrast, liability insurance looks quite different from the insurer’s point of view. Commercial (as opposed to social) insurance is a business, and this is as true of liability insurance as of any other line of insurance. Commercial liability insurers are in business to make money by selling a commodity — liability insurance. Many people (other than insurers, of course) find it quite difficult to think about insurance in this way. Compare insurance with food, for example. People need food to stay alive, but no-one thinks that it is the responsibility of commercial food manufacturers to feed the hungry. On the other hand, many people do believe, I think, that insurers have some sort of moral obligation to ensure that the social functions of liability insurance — risk spreading and compensation — are performed. In this sense, insurance is viewed rather like an essential service. For instance, even when public utilities such as gas or electricity are privately owned, the owners typically have various legal obligations to maintain supply. Analogously, one might think, liability insurers should be under an obligation to maintain supply, especially in relation to compulsory liability insurance. However, liability insurance is not a public utility. Liability insurance may be compulsory in certain circumstances, but the compulsion is on the insured, not the insurer. The engine of the liability insurance industry is the law of supply and demand. This is not to say that the liability insurance market is not regulated to prevent serious market failure; but regulation is primarily concerned with solvency and consumer protection, not with entry into and exit from the market by suppliers, or with control of the price of insurance, both of which have an important impact on the balance between supply of, and demand for, insurance products.
So-called ‘crises’ in liability insurance are, I suggest, a product of an expectation gap. Since liability insurance performs important social and economic functions, especially in relation to personal injury, people expect its price and availability to be relatively stable. Sudden reductions in supply and rises in prices defeat such expectations. They often lead to accusations of malpractice and profiteering by insurers, but those who make such accusations may fail to take due account of the iron law of supply and demand. It does not follow, of course, that insurer incompetence and mismanagement may not play a part — the final report of the HIH Royal Commission and the preliminary report of the provisional liquidator of UMP make for depressing reading in this regard. But most insurance companies are well run, and yet there are still insurance crises. So how are insurance crises to be explained?
There is obviously a link between the cost of liability and the cost of liability insurance because the cost of liability is not only one of the costs of carrying on liability insurance business, but the largest single cost. As already noted, however, the available evidence does not allow a proper assessment of the relationship between changes in liability costs and changes in insurance premiums in recent years. Even so, there is room for serious doubt that increasing claims costs are the whole explanation for the current situation. So what other factors may have played a part? One initially plausible explanation is the reduction in supply in the public liability and medical indemnity sectors respectively consequent upon the collapse of HIH and UMP. However, the prevailing wisdom is that these events were themselves symptoms of a larger systemic factor that commentators call ‘the cyclical nature of the liability insurance market’. This phrase refers to the fact that the liability insurance market experiences much greater fluctuations in the balance of supply and demand, and hence much greater fluctuations in commodity price, compared to markets in many other commodities. Whereas fluctuations in the price of certain commodities (minerals and metals, for instance) tend to be demand-led, fluctuations in the price of liability insurance appear to be predominantly supply-led. What are the features of the insurance market that give it this cyclical nature?
I want to suggest that the liability insurance market has at least three features that contribute to its cyclical nature, and hence to its propensity to experience ‘crises’. First, relative to certain other markets, the costs of entering and exiting the liability insurance market are relatively low, at least at the margin. When profits are good, it is relatively easy for existing players to increase supply and (perhaps) for new players to enter the market. Conversely, when the going gets tough, it is relatively easy for supply to be reduced. Multinationals may exit a particular market entirely for a period and return when conditions improve. Sudden changes in supply may produce equally sudden changes in price. Sudden price increases may, in turn, result in reduced demand. Whereas reduced demand would be expected — other things being equal — to trigger reductions in price, this may not occur in the insurance market. For one thing, purchasing liability insurance may be compulsory, either de jure or de facto. But even if it is not, those who react to premium increases by purchasing less liability insurance are likely to be those who need it least, because the risk that they will incur liability is lower than average — they are ‘good risks’. As a result, insurers may need to raise premiums even further in order to compensate for the fact that their portfolio of customers presents a higher aggregate level of risk than formerly because it contains a higher proportion of ‘bad risks’. Therefore, reductions in demand may, perversely, lead to further price increases, rather than reductions in price.
Secondly, market share is extremely important in the insurance market in general and the liability insurance market in particular. According to the HIH Royal Commission, in the year to December 2000, the top 20 insurers in Australia accounted for approximately 88 per cent of gross premium income, and the top five insurers accounted for half of this. Market share is important not only because of economies of scale, but also because insurance critically depends on the law of large numbers. The viability of insurance as a business depends on the portfolio of insured risks containing a good mix of low, medium and high risks. The larger an insurer’s share of the market, the more balanced its portfolio is likely to be. Insurers therefore have an incentive to compete strongly for market share. It is also the case that it is much easier for liability insurers to compete on price than on quality, at least in the retail market. Many purchasers of liability insurance probably understand relatively little about the details of the cover they are buying, and because of the complexity of the product, it is often very difficult to make meaningful comparisons between insurance products in these terms. So people are more likely to choose between rival insurers on the basis of price than by reference to the terms of the various policies on offer. Price competition is likely to be particularly aggressive when rates of return on investments are high. During such periods, insurers can afford to lower premiums in order to improve their market share even if, as a result, they suffer underwriting losses (that is, even if total premium income is less than the total cost of payouts to policy holders). However, this sort of behaviour is only sustainable so long as underwriting losses are counterbalanced by investment gains. When this ceases to be the case, the struggle for market share may take the form of consolidation through acquisitions or mergers.
Thirdly, insurance differs from most other commodities in that the major cost of supplying the commodity, that is, meeting claims, is not incurred until after the commodity has been paid for — after the premium has been received. This means that the supplier has to calculate the price of the product on the basis of its best estimate of the major cost of producing it. This gap between the time when the price must be set and the time when the cost of claims is known is called the insurance ‘tail’. All insurance has a tail. Some lines — including some lines of liability insurance — have a ‘short’ tail, but others (professional indemnity, for instance) have a ‘long’ tail. The longer the period between the date when the premium has to be fixed and the time when claims are likely to be made, the longer the tail. In order to manage the tail, insurers need to build up reserves. The longer the tail, the harder it is to calculate what those reserves need to be in order to allow for uncertainty about features of the claims environment — such as changes in the law and above-average inflation of medical costs — that may affect the number and size of future payouts. This makes long-tail liabilities extremely risky: if the insurer significantly underestimates claims, it may be extremely difficult, if not impossible (depending on the level of competition in the market), for it to make up the difference between its reserves and its liabilities by imposing short-term premium increases. For the insurer, the rational reaction to uncertainty is to set premiums sufficiently high to meet the worst-case scenario.
Reserves, then, are extremely important for the long-term stability and viability of the liability insurance market. At the same time, however, they make a significant contribution to its cyclical volatility. It is reserves that generate investment income, and it is investment income that enables insurers to compete aggressively on price at the cost of incurring underwriting losses. Equally importantly, the fact that liability insurers do not need to rely entirely on premiums for their income helps to explain why changes in premiums might not track changes in insurers’ liability costs.
However, even the strongest insurance companies can afford to incur significant underwriting losses only for so long as they are offset by investment income; and, of course, rates of investment income are largely out of the insurer’s control. Hence, the insurer’s only option, when investment income no longer offsets underwriting losses, is to raise premiums. If investment income drops significantly, premiums must increase significantly; if investment income drops suddenly, premiums must increase suddenly. Large, well managed insurance companies usually ride this roller-coaster without mishap. However, smaller or poorly managed companies sometimes come off the rails, thus reducing supply and pushing prices up even further, at least in the short term.
In summary, we can say that although there is obviously a link between the cost of liability and the cost of liability insurance, liability costs are not the only factor affecting premiums. This conclusion does not prove that increases in the cost of liability have not contributed significantly to recent increases in the cost of liability insurance in Australia. But the explanation for the conclusion does show, consistently with the available evidence, how there could have been large and sudden increases in premiums that are not directly matched by similar increases in liability costs. Liability costs are only one of the drivers of the cost of liability insurance.
In light of this conclusion, it is important to distinguish advocacy of tort reform based on a direct link between the cost of liability and the cost of liability insurance, from advocacy based on the idea that too much of society’s resources are being spent on tort liability. An advocate of reform might consistently claim that society is spending too much of its resources on the tort system without believing that the current insurance crisis is a direct result of increases in the amount spent on tort compensation. To such a person, high insurance premiums may appear to be a symptom rather than the cause of the problem — the problem being that if society spends too much on the tort system, it will have too little left to meet competing calls on its resources that should be given a higher priority. From this perspective too, the problem is not legal but financial. Law reform is a means to an end — namely, the reallocation of resources away from the tort system to other valued activities — but not an end in itself.
There is, however, another important strand in contemporary Australian debates, influentially expounded in an oft-cited speech given by New South Wales Chief Justice James Spigelman early in 2002 entitled ‘Negligence: The Last Outpost of the Welfare State’. This strand focuses on the balance struck by tort doctrine between the interests we all share in health and safety on the one hand, and freedom of action on the other. As its title suggests, the thesis of Spigelman’s article was that tort doctrine had become too favourable to claimants and that there was an urgent need for what he called ‘principled’ reform of the law to redress the balance. This is not an argument about how much tort law costs, but rather about the normative values according to which the law does (and should) regulate relations between individuals and social groups.
These two distinct strands in debates about tort law reform and the insurance crisis came together in the terms of reference of the Review. The terms of reference begin strongly with an assertion that the law of negligence has become ‘unaffordable and unsustainable as the principal source of compensation for those injured through the fault of another.’ However, they are also peppered with phrases such as ‘self-assumption of risk’, ‘allowing individuals to assume risk’ and restriction of ‘the circumstances in which a person must guard against the negligence of others’. The idea of law reform as a means to an end is reflected in the recommendation of the Trowbridge Report that the terms of reference should ‘be as specific as possible regarding the desired outcomes’. Thus they assign the Review ‘the objective of limiting liability and quantum of damages arising from personal injury and death’.
A so-called ‘Panel of Eminent Persons’ (‘the Panel’) was appointed early in July 2002 to conduct the Review. The Trowbridge Report had recommended that the Panel should include ‘community representation’ and that
[t]he panel could comprise four people (say a retired judge, an academic lawyer, a practical barrister and, say, a pragmatist with a social sciences background) with the time to devote and the knowledge and interest to tackle the challenge. They would need support from a talented and enthusiastic legal researcher.
The Panel consisted of a sitting appeal court judge (as chair), an academic lawyer, a clinical professor of surgery and a senior member of local government. It was supported by a predominantly legal staff of public servants and a few secondees from the private legal sector — about 15 in total. The Panel started work late in July 2002, submitted its first report to the Commonwealth Assistant Treasurer and Minister of Finance, Senator Helen Coonan, on 30 August, and its second and final report on 30 September. The Review received about 100 written submissions and held consultations with judges, practising and academic lawyers, psychologists, government agencies, such as the ACCC, and stakeholder representatives, such as the APLA, the Insurance Council of Australia (the peak body of the insurance industry), the Law Council of Australia (the peak body of the legal profession) and so on. The consultations took up about two weeks of the available time. The rest was spent in intensive research, a great deal of discussion amongst the members of the Panel and with its support team, the formulation of proposals and the writing of the Review Report.
In order to understand the Review and its place in the larger picture of tort law reform, it is necessary to appreciate the political environment in which it was conducted. The Review was only one limb of a multi-pronged government reaction to a situation that was believed by some to have the potential to cause major social disruption as a result, for instance, of cessation of a range of community activities and long-term unavailability of certain types of medical services in some areas of the country. The terms of reference were limited to liability for negligence, and to personal injuries and death, reflecting the fact that the Review was seen as part of the answer to a specific problem rather than as a general review of liability law. The terms of reference reflected the immediate political and social situation in other respects, too. On the one hand, they gave the Panel a very broad remit to consider the principles of liability and assessment of damages for negligently-caused personal injuries and death. On the other hand, however, within that broad remit they also focused on a number of specific issues that reflected immediate political concerns and hinted at government priorities in reacting to the various pressures to which it was subject. For instance, one of the terms of reference related to the liability of local government authorities, who were concerned about the potential impact on local government finances of the abolition of the so-called ‘highway immunity’ by the High Court in Brodie v Singleton Shire Council. Another clause of the terms of reference dealt with ‘professional negligence matters (including medical negligence)’, and the clause dealing with limitation periods was known to be of particular interest to doctors, especially obstetricians (as well as to medical indemnity insurers, of course).
Concerns about the impact of increases in liability insurance premiums on community life, especially in rural and regional areas, were reflected in an instruction to develop and evaluate options for exemptions or limitations of liability in favour of ‘not-for-profit organisations’ including ‘community service and sporting organisations’. Concerns about the future of the adventure sports and tourism industries found expression in an instruction to review the operation of the Trade Practices Act 1974 (Cth) and, in particular, a proposed amendment to the Act designed to exempt providers of recreational services from a prohibition on contractual exclusion of liability for breach of a statutorily implied obligation of reasonable care. The Panel was required to report in two stages: a first report by 30 August 2002 and a final report by 30 September 2002. The matters on which it was required to report first — professional liability, limitation periods, the liability of not-for-profit organisations and the Trade Practices Act 1974 (Cth) — were arguably the most politically sensitive of the issues covered by the terms of reference.
These and other aspects of the terms of reference presented the Panel with a major challenge. On the one hand, it was clear that the brief was not to conduct a general review of personal injury law in the relatively unconstrained way that a standing law reform commission might be asked to do. In this, as in many other respects, the Panel was in a very different position from the Law Commission for England and Wales which, over a period of eight years, produced an extensive series of consultation papers and reports on various aspects of the law of damages, especially damages for personal injury and death. In the view of many academic commentators, such a long-term programme of work represented an ideal to which the Review obviously could not aspire. For these critics, the prospect that the Panel’s report might form the basis of legislative action was deeply troubling. The Review was focused, as the Trowbridge Report had recommended, on developing a set of proposals directed to a stated policy objective of limiting liability and damages for negligence. Proposals that did not further this objective, it was implied, would not fall within the terms of reference, however desirable or justifiable they might seem judged according to some other objective or criterion. In this way, the Review could be interpreted as an exercise in ‘technical’ law reform, with the job of the Panel being to suggest technically feasible ways of furthering the policy objective so clearly stated in the terms of reference.
On the other hand, the objective of limiting liability and quantum of damages for negligence is so broad, and open to so many different interpretations and degrees of implementation, that it was inevitable that the Panel would have to make a great many non-technical choices between options for change that were, in a technical sense, equally feasible. This is why it is so important for those reading the Panel’s report (as it was for the members of the Panel in conducting the Review) to take account of the political environment surrounding the establishment and operation of the Review. In this regard, various aspects of the Review deserve to be specifically noted, including the fact that the reform process was driven by Treasury Ministers, rather than Attorneys-General. This indicates that the underlying problem was seen as being primarily economic rather than legal.
Another aspect of the Review which should be highlighted is the fact that the Panel was given little more than two months to conduct the Review. This limited time frame could be interpreted as suggesting that governments saw the exercise partly as providing a way of being seen to be doing something about the insurance crisis. It is worth remembering that although the federal government took the lead in promoting a review of negligence law, the Review was jointly sponsored by all nine Australian governments. The liability insurance industry is national, and although New South Wales and Victoria between them have the lion’s share of tort claims and litigation, governments in all jurisdictions had a political interest in dealing with the insurance crisis. However, it seems unlikely that anyone believed that major legal reform on a nationally uniform scale could be achieved overnight.
In this regard, it is worth noting two further facts. Personal injury law has for many years been fragmented by various statutory regimes dealing, for instance, with transport accidents and workplace injuries. There are significant substantive differences between such regimes, and some involve special institutional arrangements (as in the case of the dust diseases regime in New South Wales). The other fact to note is that the Review was by no means the first step in a process of limiting liability for personal injuries. For instance, a bewildering variety of statutory provisions limiting quantum of damages has existed for years in Australia, especially in New South Wales. More immediately to the point, perhaps, New South Wales was engaged throughout 2002 in a particularly active (not to say frenetic) programme of legislative changes to personal injury law which, at least until the Review was well under way, was pursued more or less independently of the more coordinated processes supported by the federal government. The attitude of the New South Wales government appeared to be that the insurance crisis presented it with unique, and uniquely urgent, problems — solutions to which could not await the formulation of a national response. Against this background, it is not implausible to interpret the Review as an expedient to deal with an immediate problem rather than as part of a larger strategy of national reform and rationalisation of personal injury law.
There is no doubt that certain stakeholders — such as the Insurance Council of Australia — were strongly in favour of nationally uniform legislation. The federal government also used and encouraged the rhetoric of national coordination. Once again, this approach can be traced back to the Trowbridge Report, which distinguishes areas in which it is desirable that any reform measures be nationally uniform from those in which ‘consistency’ between jurisdictions would be adequate, and from those in which neither uniformity nor consistency is ‘critical’. Yet it is hard to imagine that any of the politicians involved in the process seriously believed that the Review would or could have led to truly collaborative legislative action in all jurisdictions in the near future or indeed at all. Given the multifarious differences between existing statute-based tort law in the various states and territories, any attempt to achieve consistency, let alone uniformity, would have required an implausible amount of ground-clearing even before construction of a new legal regime could have begun. Even so, the federalist case for diversity seems quite weak in many areas of personal injury law, and although not instructed to do so by its terms of reference, the specific recommendations of the Review Report were developed with an eye to promoting the values of uniformity and consistency across jurisdictions and are framed by a general recommendation for nationally uniform legislative implementation.
The time scale of the Review was, unsurprisingly, the subject of much criticism and it placed significant constraints on the amount of detail that the Panel could provide both in its recommendations and in the reasoning used in their support. It was suggested to the Panel more than once that an extension of time should be sought. However, it was clear that no extension would have been granted: before the Panel even started work, a ministerial summit to consider its report had been scheduled for October 2002. One can only assume that neither the authors of the Trowbridge Report, nor the ministers who decided to accept their recommendation for a review of the law of negligence, had any real idea of the potential magnitude of the task.
Academic observers tended to view the Panel’s job in terms of their traditional patterns of research and writing, and to conclude that the whole process lacked intellectual credibility. Certainly, although the Panel had excellent research support, it could not be said that the conduct of the Review was research-led. The combined experience and knowledge of the two legal members of the Panel provided the essential foundation of the exercise. The suggestion that the Panel should have investigated the assumption made by many (although, as we have seen, neither by Trowbridge nor the Panel’s terms of reference) that limiting liability and the quantum of damages for negligence would lead to a fall in liability insurance premiums, was not only politically naïve but also impractical in light of the resources of time and staffing made available to the Panel. What the sponsors of the Review wanted were practical and focused proposals to deal with acute social and political problems. The prime audience for the Panel’s report were politicians, not lawyers or academics. The qualities most demanded of the Panel were not erudition and painstaking analysis but decisiveness and what the great realist jurisprude, Karl Llewellyn (speaking of the judicial task), called ‘situation sense’. It is against such criteria that the Panel’s report should be assessed and not in terms of the more leisurely virtues of the academy or the law reform commission.
The membership of the Panel also reflected the political climate in which it was established. Law reform bodies are typically staffed wholly or predominantly by lawyers. Only two of the members of the Panel were trained lawyers. The other two members were identifiable with two of the main stakeholder groups (local authorities and the medical profession), perhaps in response to a recommendation in the Trowbridge Report that the Panel contain ‘community representation’. The presence of the non-legal members was a cause of much criticism, especially (and perhaps predictably) by lawyers. Indeed, written submissions from certain key legal representative bodies contained inappropriately personal attacks on some members of the Panel, as well as more measured general comment on the Panel’s composition. As was to be expected, the non-legal members of the Panel were conscientious, hard-working and open-minded. They made significant contributions to the Panel’s consideration of issues relevant to their experience and expertise.
To the extent that criticism of the composition of the Panel rested on some idea that reform of personal injury law is exclusively lawyers’ business, it should be rejected. At the same time, lack of legal expertise limited the extent to which the non-legal members could participate in debates about (and formulation of) the Panel’s recommendations and in the preparation of the Review Report. One undesirable result was that the burden of this work fell almost exclusively on the two legal members. This made the shortage of time more significant than it might otherwise have been and — just as importantly — deprived the Panel’s deliberations on many issues of a certain degree of diversity of informed opinion and debate.
Because of its political salience, the Review attracted much more media attention than the typical law reform exercise. A high point, perhaps, was the edition of Australia Talks Back on Radio National devoted to the Panel’s first report. Of course, the insurance crisis and the events that led up to the establishment of the Review filled many columns in the general press, as well as in more specialised publications such as The Australian Financial Review and various professional and trade journals. Nevertheless, the amount of interest in the Review Report itself was, perhaps, less to be expected. The role of the media in influencing and creating understandings of law and legal institutions is a topic that deserves much more scholarly attention than it has hitherto received. Coverage of the insurance crisis, and of the Review and the various legislative reactions to it, would provide an excellent case study.
All this shows, I think, that the Review should not be understood as a typical law reform exercise. The prime audience for the Review Report was made up of ministers and politicians, not lawyers within and outside government. The Panel had to try to make complex legal concepts and ideas accessible to people without any legal training. The physical location of the Panel’s offices in the Treasury building in Canberra was not without significance. The Review Report should, I suggest, be read as expert policy advice to government or, perhaps, as instructions to a legislative drafter, not as a set of finely-tuned recommendations for the reform of ‘lawyers’ law’.
However, it should also be said that the concepts of ‘technical law reform’ and ‘lawyers’ law’ are themselves lawyers’ conceits. Politics involves the exercise of power, and all laws are exercises of power, whether by legislatures, executives or courts. Hence, all laws are political and, potentially at least, of interest to politicians and the political process. Politicians are happy to leave to law reform bodies only matters about which they care little or which, on the contrary, they find too hot to handle. In the current situation in Australia, the law of negligence falls into neither of these categories. Australian governments currently care a lot about negligence law, and however ‘hot’ it might be, they feel a political need to address it. The Review was established not to sideline negligence law and remove it from the political arena, but rather to provide expert legal input into the political policy-making process. The Review was inevitably informed by the political purposes for which it was established.
The loudest critics of the Review (outside the academy, anyway) have been plaintiffs’ lawyers’ representatives. Much of their criticism has been couched in the language of legal ‘rights’ which, if taken seriously, would stifle any departure from the status quo. It is certainly possible and entirely permissible to object to the Review Report’s recommendations on the ground that they worsen the position, under tort law, of injured people relative to injurers. It is not plausible to dress this up, as critics often do, as a legal rather than a political objection. Some people — whom I earlier dubbed ‘compensationists’ — oppose legislative amendment of judge-made tort law on the basis that it will upset a carefully crafted judicial balance of rights and obligations by introducing discordant political compromises into the common law, thus destroying its ‘coherence’. But coherence is relative to the law’s goals and lawyers too often reject changes in the law as creating incoherence when what they really object to are the goals which the changes are designed to promote. In this context, it is worth noting that from the mid-19th to the mid-20th centuries, legislative interventions in tort law were predominantly pro-claimant. It is only in the last 20 years that legislatures have started changing tort law in a pro-defendant fashion.
Some commentators have argued that the Review was unnecessary and undesirable because the High Court had already readjusted the balance of rights and obligations in personal injury law in the way the Panel recommended. Even if this is accepted, I believe that because of their social and economic significance, it is much more desirable that major changes in basic principles of tort law be the result of political and legislative, rather than judicial, activity.
This is not the place to devote detailed attention to the substance of the Review Report. However, a few comments may be in order.
It was noted earlier that various strands in recent debates were reflected in the terms of reference and, therefore, inevitably in the Review Report. A careful reading of the Review Report should reveal the dominance in the Panel’s thinking of the idea that tort law is essentially a system of rules and principles of interpersonal responsibility for harm-doing. Adoption of this perspective resulted from a recognition that the Panel was not free to question the assertion of unaffordability in the terms of reference, and that it lacked the resources to investigate either the link between the operation of the personal injury liability system and volatility in the liability insurance market, or the likely impact of its recommendations on the cost of liability insurance. Certainly, the normative proposition that, in terms of responsibility ideas, tort law was too favourable to claimants was controversial, but at least it did not rest on contentious empirical assumptions. It is concerned with the inner normative logic of the law rather than with its external social effects. Putting the same point slightly differently, the Panel interpreted its task as being concerned with legal doctrine, not social engineering.
I also noted earlier that the terms of reference left the Panel with generous leeways of choice in seeking to promote the objectives of the Review. Even so, because the terms of reference were so explicit about those objectives, in some contexts the Panel felt it possible and desirable — especially given the time scale of the Review — to appeal expressly to the objectives of the terms of reference to justify particular recommendations without reference to the preferences or opinions of Panel members. The Panel also felt able to reject certain proposals for change on the basis of their inconsistency with the terms of reference, regardless of whether those proposals might have been thought to represent desirable ‘improvements’ to the law. In other words, the objectives stated in the terms of reference were relied upon to define the parameters of the Panel’s freedom of choice.
The proposals on psychiatric injury deserve some attention not because of their content, but because the circumstances in which they were formulated cast an interesting light on processes of legal development and change. The High Court’s decision in Tame v New South Wales was handed down at the very time the Panel was considering this aspect of the law. It seemed clear that whatever the Panel recommended, it would have to take account of what was undoubtedly a decision of major significance. But how was it to deal with a case in which the seven Justices gave six substantial judgments running to 90 pages in total in the law reports supported by more than 300 footnotes? In one respect there was no problem because the decisions on the two appeals, dealt with together by the Court, were unanimous. In addition, five of the six judgments were similar in their broad thrust. The problem the Panel faced, however, brings into sharp focus the difference between judicial activity — even in the highest courts — and legislative activity. The Panel conceived its task essentially as being to develop policy instructions to legislative drafters. By contrast, courts in the common law tradition — even appellate courts — see their prime function as the resolution of disputes. The High Court’s only obligation was to resolve the appeals before it either in favour of the plaintiff or the defendant. It had no obligation to provide a single, rule-like basis for that resolution. Individual judges of appellate courts have no obligation — and typically behave as if they have no obligation — to cooperate with their colleagues to produce anything like policy instructions to legislative drafters or even, it often seems, to give clear guidance to lawyers and lower courts for the resolution of future disputes in or out of court. They need not avoid inconsistencies between their various judgments or adopt the same conceptual approach. So long as each maps out a path to the result, it matters not that each path is different. In Tame the paths ran parallel for much of their length, but no two were the same and there were some significant differences between them.
The law governing liability for psychiatric injury is notoriously unstable and complex. The decision in Tame seemed to offer useful raw materials for a ‘restatement’ of the law that would further the objectives of the terms of reference. Out of these materials the Panel needed to fashion a relatively straightforward and coherent set of principles that ironed out creases in the composite fabric of the various judgments and which could be used as the basis for legislation. Inevitably, this process required the Panel to make decisions about which strands in the reasoning to emphasise and which to marginalise. The Panel saw its task not in terms of giving effect to the High Court’s decision, but rather as being to take the decision as a starting point for developing a robust, if controversial, set of proposals that would promote the objectives of the Review. Here, as elsewhere, the Panel was guided by the idea of tort law as a set of principles of interpersonal responsibility.
Reports and recommendations of law reform bodies often meet with deafening silence from government. Although the reaction to the Review by the various Australian governments has been far from uniform, the Review Report has certainly not been ignored. It was the subject of a joint ministerial meeting in October 2002 at which the chair of the Panel spoke, and there was another meeting in April 2003 to consider progress with implementation. In New South Wales, Queensland and Western Australia, for instance, various recommendations of the Panel have been enacted more or less verbatim. A lesson that could be drawn from this is that the more politically salient the activities of law reform bodies, the more likely those activities are to provoke a legislative reaction. The aftermath of the Review reinforces the point that it can only be understood in the context of the political environment which produced it. By mid-2003 the political focus and the attention of the media had shifted to liability for purely economic loss and (non-medical) professional indemnity insurance.
In the current Australian political climate, what can be said about the prospects for more radical reform of tort law? Some people saw the Review as a lost opportunity to recommend abolition of tort and the introduction of a no-fault scheme. Indeed, from one point of view, the most effective way of limiting liability and quantum of damages for negligence would be to abolish tort law and replace it with a no-fault administrative compensation scheme. But on no reasonable reading of the Panel’s terms of reference could such a proposal have been considered to fall within its remit. Even so, some of the concerns that led to the establishment of the Review — the amount of resources the tort system consumes, and the line it draws between those injured as a result of another’s fault and those not — are amongst those that support the case for abolition of the tort system of compensating for personal injuries. Where the two perspectives come apart is in the role they give tort law as a regulatory mechanism to promote health and safety. Whereas radicals tend to downplay the regulatory potential of tort law, moderates tend to attribute to it a significant regulatory role and to support limitation of the incidence and quantum of liability partly on the basis that strongly pro-plaintiff tort law encourages excessive investment in health and safety and excessively discourages socially desirable (albeit risky) activities — in short, ‘overkill’ or ‘over-deterrence’.
It seems to me that this is the angle that proponents of the abolition of tort and its replacement by a no-fault scheme (however funded) need to exploit in order to turn dissatisfaction with the tort system into pressure for its abolition. On the one hand, the evidence we have about the positive regulatory impact of tort law is patchy and inconclusive. On the other hand, as we have seen, the relationship between tort law and the insurance crisis — and hence the negative effects on social activities attributed to it — is by no means clear. On the whole, while it seems reasonable to think that the imposition and possibility of tort liability has some impact on people’s conduct, the precise nature and magnitude of that impact is largely speculative and there is good reason to think that tort law is, at best, a blunt regulatory instrument. In this light, the critical question that must be posed is whether the huge cost of delivering tort compensation (consistently estimated, in aggregate, as high as 40 per cent or more of the total cost of the tort system, and relatively much higher than the cost of delivering no-fault compensation) is worth the ‘benefits’ that would be lost in the move to a no-fault system, namely the attribution of responsibility with whatever its associated incentive effects may be. Such a reorientation of public debates away from compensation and towards deterrence offers some hope of further progress towards the goal of abolition of tort. However, particular attention would need to be paid to the potential use of the tort system to mount mass actions against large corporations. Even convinced abolitionists may have to concede that this form of consumer activism has some value. Imagining and developing alternative means of curbing the excesses of corporate behaviour would be an important part of advocacy for a no-fault system. Abolitionists would also be wise to pay serious attention to frequently expressed concerns about the potential effects of abolition of tort on doctors’ incentives for safety.
It is important to acknowledge that managing the shift to a no-fault system would create problems and complexities of its own. One particularly thorny issue is the relationship between any such no-fault scheme and the sickness and disablement elements of the general social security system. Another related issue is whether the scheme would extend to diseases as well as accidents. The Australian Woodhouse proposals originally encompassed diseases, but the scheme that eventually went before Parliament was limited to accidents. All these are issues of the greatest significance with which abolitionists must grapple if they are to be taken seriously.
Pending its demise, what attitude should abolitionists take to tort law and debates about its future? I wish to answer this question indirectly. In Gleaner Co Ltd v Abrahams the Privy Council had to consider how general damages for defamation ought to be calculated. In an attempt to control such awards, English courts have drawn an analogy between damages for non-pecuniary loss in personal injury cases and general damages for defamation. The Jamaican Court of Appeal rejected this analogy, and one issue before the Privy Council was whether this rejection constituted legal error. In denying that it did, Lord Hoffmann said:
Once it is appreciated that the awards [for non-pecuniary loss in personal injury cases] are not paid by individual defendants but by society as a whole or large sections of society, there are also considerations of equity between victims of personal injury which influence the level of general damages. Compensation, both for financial loss and general damages, goes only to those who can prove negligence and causation. Those unable to do so are left to social security: no general damages and meagre compensation for loss of earnings. The unfairness might be more readily understandable if the successful tort plaintiffs recovered their damages from the defendants themselves but makes less sense when both social security and negligence damages come out of public funds. So any increase in general damages for personal injury awarded by courts only widens the gap between those victims who can sue and those who cannot.
This is a clear statement of one of the classic abolitionist criticisms of the tort system. It led to a vigorous public email exchange (in the Obligations Discussion Group hosted by the University of Western Ontario) between Jason Neyers and Harold Luntz (amongst others). Neyers objected to Lord Hoffmann’s appeal to what Neyers called ‘distributive justice reasoning’. Leaving aside the contentious issue of what this term might mean, in essence Neyers objected that Lord Hoffmann’s approach was inconsistent with a responsibility-based conception of tort law. In reply, Luntz argued (in effect) that it should be relevant to the assessment of damages for non-pecuniary loss in a personal injury case (i) that the tort system is funded in a regressive way (because damages are earnings-related but liability insurance premiums are not) and (ii) that money spent on the tort system is money not available for other social purposes (such as public health care for the poor).
Radicals should take heart from the fact that five senior common law judges are prepared to take seriously arguments against the tort system that have been central to debates about personal injury law for 40 years or more. I doubt that a strategy of marginalising popular discontent with tort law on the ground that it is concerned with the wrong issues is the best one for abolitionists. As I have already argued, a better approach would be to exploit dissatisfaction with the cost of the tort system and its apparent inequities by focusing on its shortcomings as a regulatory instrument. In other words, perhaps abolitionists should stop standing aloof from current debates about tort law and, instead of treating them as part of the problem, start thinking of them as an opportunity. From this perspective, the Review might look more like the beginning of the end rather than just another episode in the sad history of radical tort reform in the past 40 years. If people can be convinced that the tort system is as bad as the radicals believe, and much worse than is popularly thought, they may be prepared to support not just its limitation, but its total liquidation.
There is one lesson above all others that radicals should take from the Review and that concerns the importance of political engagement. It is only by exploiting the present situation to get the cause of no-fault systems out of the law reviews and onto the political agenda that radicals have any real prospect of making significant gains. The Review has had an impact because it was politically salient. Advocacy of a no-fault system could also have an impact if it was politically salient. In fact, there is one aspect of current debates that may provide radicals with the thin edge of a large wedge. The Legal Access Reform Group of the Australian Health Ministers Advisory Council supports removal of long-term care costs from the tort system and their provision through a statutory administrative process. The details of the proposal are sketchy and its scope is limited to the medical indemnity context. Nevertheless, experience in Great Britain and Australia suggests that the medical profession may provide powerful political allies in the cause of promoting no-fault alternatives to the tort system. Spotting and exploiting such opportunities is essential for achieving truly radical legal reforms.
[∗] Law Program, Research School of Social Sciences, The Australian National University.
 For a recent illustration of all these qualities, see Harold Luntz, ‘Medical Indemnity and Tort Law Reform’ (2003) 10 Journal of Law and Medicine 385.
 The idea of ‘reform’ will be further discussed in Part II.
 Luntz, above n 1, 393.
 The Review was conducted by the so-called ‘Panel of Eminent Persons’ consisting of Justice David Ipp, Associate Professor Donald Sheldon, Mr Ian Macintosh and myself. The Panel’s final report — incorporating its first report, published in August 2002 — was presented in September 2002: Panel of Eminent Persons, Review of the Law of Negligence: Final Report (2002) (‘Review Report’).
 Peter Cane, ‘Consequences in Judicial Reasoning’ in Jeremy Horder (ed), Oxford Essays in Jurisprudence: Fourth Series (2000) 41.
 See Peter Cane, ‘Distributive Justice and Tort Law’  New Zealand Law Review 401.
 See Peter Cane, Atiyah’s Accidents, Compensation and the Law (6th ed, 1999).
 See Patrick Atiyah, The Damages Lottery (1997).
 Guido Calabresi, The Cost of Accidents: A Legal and Economic Analysis (1970). For a history of no-fault debates in the United States, see Richard Gaskins, ‘The Fate of “No-Fault” in America’ (2003) 34 Victoria University of Wellington Law Review 213.
 Ibid 14.
 Ibid 26.
 Terence Ison, The Forensic Lottery: A Critique on Tort Liability as a System of Personal Injury Compensation (1967).
 Ibid ix.
 Ibid 89.
 Patrick Atiyah, Accidents, Compensation and the Law (1970) 593.
 Ibid 591.
 New Zealand, Royal Commission of Inquiry into Compensation for Personal Injury in New Zealand, Final Report (1967).
 See generally Stephen Todd et al (eds), The Law of Torts in New Zealand (3rd ed, 2001) 82–4.
 National Committee of Inquiry, Compensation and Rehabilitation in Australia (1974). Harold Luntz helpfully analysed the report in Compensation and Rehabilitation: A Survey of the Report of the National Committee of Inquiry into Compensation and Rehabilitation in Australia and the National Compensation Bill 1974 (1975). See also Harold Luntz, ‘Looking Back at Accident Compensation: An Australian Perspective’ (2003) 34 Victoria University of Wellington Law Review 213.
 United Kingdom, Royal Commission on Civil Liability and Compensation for Personal Injury, Report (1978).
 See Motor Accidents (Compensation) Act (NT).
 See Transport Accident Act 1986 (Vic).
 New South Wales Law Reform Commission, Accident Compensation: A Transport Accidents Scheme for New South Wales, Report No 43 (1984).
 Patrick Atiyah, ‘Personal Injuries in the Twenty First Century: Thinking the Unthinkable’ in Peter Birks (ed), Wrongs and Remedies in the Twenty-First Century (1996) 1.
 Atiyah, The Damages Lottery, above n 8, 157–8.
 Ibid 141 (emphasis in original).
 Ibid 142.
 Ibid 165.
 See, eg, Senate Economics References Committee, Parliament of Australia, A Review of Public Liability and Professional Indemnity Insurance (2002) ch 1 (‘Senate Review’).
 The ACCC has a brief to monitor premiums for public liability and professional indemnity insurance in the light of the various steps being taken to limit increases in insurance costs. The first of four planned reports — ACCC, Public Liability and Professional Indemnity Insurance: Monitoring Report (2003) — contains a lot of data, but it is not at all easy to interpret. A few general points are worth making. Firstly, although the number of public and professional liability claims increased between 1997 and 2002, the frequency, relative to the number of policies issued, fell: at 135. Nevertheless, insurers cited ‘an increasingly litigious society’ as one of the most significant factors in explaining increases in payouts and premiums: at 60. Secondly, the statistics about the cost of professional indemnity insurance do not distinguish between personal injury and pure economic loss claims: at 26. Thirdly, although information is provided about claim numbers, their frequency and the average size of payouts, these categories of information are not aggregated to produce statistics about (changes in) the total cost of liability payouts or the relationship between total premium income and total payouts. Fourthly, there is limited information about the underwriting cycle — that is, the time lag between the setting of premium levels for a particular year and the calculation of underwriting results for that year. It is obvious that premium levels for any particular year — say calendar 2002 — cannot reflect actual underwriting results for that year or even for the year before. It is therefore very difficult to trace the connection between changes in premiums and changes in payouts. Fifthly, the ACCC’s survey of insurers indicated that the most important pricing method was to set premiums by reference to desired profit levels, not by reference to the expected cost of claims: at 52. Finally, the ACCC comments expressly on the failure of insurers to provide all the information requested: at 4.
 Trowbridge Consulting Ltd, Public Liability Insurance: Practical Proposals for Reform (2002) Commonwealth Treasury <http://www.treasury.gov.au/documents/314/PDF/plr.pdf> (‘Trowbridge Report’).
 Ibid 59.
 It should be noted that the Trowbridge Report was concerned with public liability insurance. So far as medical indemnity is concerned, the provisional liquidator of UMP reported that there were ‘“spikes” or significant increases in claims in both number and quantum’ in New South Wales in 1998 and 2001: Deloitte Touche Tohmatsu, United Medical Protection Ltd: Preliminary Report into Solvency and Position as at 3 May 2002 (2002) 10 <http://
www.unitedmp.com.au/0/0.12/0.12.3.pdf> (the latter, apparently, in anticipation of the commencement of the Health Care Liability Act 2001 (Cth), which imposed various limits on medically-related personal injury claims). This was one of the factors that led to the collapse of UMP. Failure to build up adequate reserves to meet future claims was another. What the spikes as such appear not to have precipitated were immediate large premium increases.
 Trowbridge Report, above n 32, iv.
 Ibid 26–7.
 Ibid 12.
 Peter Cane, Tort Law and Economic Interests (2nd ed, 1996) 480–1.
 See, eg, Tony Honoré, Responsibility and Fault (1999) 85–7. Whether or not courts should take account of the availability of liability or loss insurance in fashioning liability rules is a highly controversial issue.
 Atiyah, Accidents, Compensation and the Law, above n 16, 283.
 The government now appears to be assuming the role of what Chief Justice James Spigelman has felicitously called the ‘reinsurer of last resort’: Chief Justice J J Spigelman, ‘Negligence: The Last Outpost of the Welfare State’ (2002) 76 Australian Law Journal 432, 434. See, eg, the range of schemes now provided by the government: Department of Health and Ageing, Commonwealth, Medical Indemnity: What the Changes Mean for You (2003) <http://
 Of course, licensing and capital adequacy requirements affect ease of market entry, but this is not their prime function. See also Senate Review, above n 30, [2.24]–[2.29].
 See especially Commonwealth, HIH Royal Commission, The Failure of HIH Insurance (2003) vol 2, ch 15 (‘HIH Royal Commission’).
 Deloitte Touche Tohmatsu, above n 34.
 In PricewaterhouseCoopers, Report to the Insurance Issues Working Group of Heads of Treasuries, Actuarial Assessment of the Recommendations of the Ipp Report (2002), the figure of 65 per cent is used to represent the notional proportion of public liability insurance premiums spent on meeting claims (that is, money spent on both compensation and costs). This forms the basis of their calculations of the likely impact of the Review on premiums.
 See above n 31.
 Another frequently-mentioned factor is 11 September 2001. It is unlikely to have contributed directly to the current situation in Australia. Indirectly, it is said to have resulted in significant increases in the cost of reinsurance. But it is worth noting that the Association of British Insurers told the Office of Fair Trading (‘OFT’) in Great Britain that any increase in the cost of liability insurance in the UK as a result of increases in the cost of reinsurance was ‘negligible’: OFT, The UK Liability Insurance Market — Summary of Key Findings (2003) [4.19]. The Australian evidence is conflicting: Senate Review, above n 30, [2.4]–[2.8].
 The OFT concluded that none of the general barriers to entry — licensing, capital adequacy requirements, and so on — ‘seem to place a substantial limit on entry’: OFT, above n 47, [2.13]. In Australia, the Productivity Commission recently concluded that in the Australian public liability market barriers to entry and exit are low: Productivity Commission, Public Liability Claims Management: Research Report (2002) 49.
 In 2001 two commercial insurers (including St Paul, the world’s largest medical malpractice insurer) withdrew from the New South Wales medical indemnity market, as a result of which no commercial insurer was writing this class of business in New South Wales. (In fact, St Paul withdrew from the medical malpractice insurance market worldwide.) It is said that following the collapse of HIH, the number of professional indemnity insurance providers in Australia fell from 33 to five: David Parken, ‘Driving Insurance Liability Reforms’, The Australian (Sydney), 21 August 2003, T16.
 HIH Royal Commission, above n 43, [4.1].
 The importance of market share can, conversely, also explain lack of competition and relatively high premiums in specialised sectors of an insurance market where there are relatively few potential customers. Not-for-profit organisations may have been victims of this effect. Some form of ‘risk-pooling’ may ameliorate this problem: Senate Review, above n 30, [4.55]–[4.65].
 According to the Productivity Commission, public liability premiums fell by about 35 per cent between 1993 and 1998: Productivity Commission, above n 48, 22.
 HIH Royal Commission, above n 43, [4.2.1].
 This is why professional indemnity insurers prefer to write claims-made policies rather than occurrence-based policies. Under an occurrence-based policy, claims can be made at any time in respect of events that occurred during the policy period, however long before the date of the claim. Under a claims-made policy, claims can be made during the policy period in respect of events that occurred within a specified period — for example, since the date when the insured first bought a policy of the relevant type from the insurer. Claims-made policies shift risk from the insurer to the insured by specifying a period during which the claim-generating events must have occurred and excluding claims in respect of events occurring after the end of the policy period. (Some of the advantage to the insurer of claims-made policies is removed by s 54 of the Insurance Contracts Act 1984 (Cth), as interpreted in FAI General Insurance Co Ltd v Australian Hospital Care Pty Ltd  HCA 38; (2001) 204 CLR 641.) This explains why the medical profession is so keen on very short limitation periods. The limitation period effectively sets the period for which a professional is at risk of being sued after retiring from practice. The longer that period, the longer the professional has to maintain active, claims-made insurance cover after retirement (‘run-off cover’).
 This explains why insurers are more concerned that the law should be certain and stable than that it should have any particular content. The more uncertain the law is, and the more subject it is to retrospective change by judicial decision, the harder it is for insurers to estimate future claims costs.
 ‘Under-provisioning’ was the major factor contributing to the collapse of both HIH and UMP.
 It is not clear how significant a factor this has been in the current Australian situation: Senate Review, above n 30, [2.9].
 The insurance industry is notoriously secretive and statistical information about liability insurance is in very short supply: ibid ch 5. This led the OFT to comment: ‘the explanations given by insurers for recent increases in premiums have not been as detailed or rigorous as we would have expected. This may be a reflection of the limited role of actuaries in underwriting decisions’: OFT, above n 47, [5.12]. Actuarial input is particularly relevant to calculating likely claims costs. If the OFT’s speculation about the role of actuaries is correct, it provides another explanation of why the relationship between claims costs and premiums might be an indirect one.
 Robert Debus, ‘Tort Law Reform in New South Wales: State and Federal Interactions’  UNSWLawJl 47; (2002) 25 University of New South Wales Law Journal 825, 825–7.
 Spigelman, above n 41.
 Review Report, above n 4, ix.
 Trowbridge Report, above n 32, 27.
 Review Report, above n 4, ix.
 Trowbridge Report, above n 32, 27.
 Review Report, above n 4, xiii.
 The whole operation was distinctly less amateurish than the tone of the Trowbridge recommendations suggests was contemplated by their authors.
  HCA 29; (2001) 206 CLR 512.
 Review Report, above n 4, x.
 Ibid x–xi.
 Ibid x.
 Law Commission for England and Wales, Structured Settlements and Interim and Provisional Damages, Report No 224 (1994); Law Commission for England and Wales, Personal Injury Compensation: How Much Is Enough: A Study of the Compensation Experiences of Victims of Personal Injury, Report No 225 (1994) (this is a major empirical study about the ‘adequacy and effectiveness’ of personal injury damages); Law Commission for England and Wales, Aggravated, Exemplary and Restitutionary Damages, Report No 247 (1997); Law Commission for England and Wales, Liability for Psychiatric Illness, Report No 249 (1998); Law Commission for England and Wales, Damages for Personal Injury: Non-Pecuniary Loss, Report No 257 (1999); Law Commission for England and Wales, Damages for Personal Injury: Medical, Nursing and Other Expenses; Collateral Benefits, Report No 262 (1999); Law Commission for England and Wales, Claims for Wrongful Death, Report No 263 (1999); Law Commission for England and Wales, Limitation of Actions, Report No 270 (2001).
 The original proposal would have given only about a month!
 See generally Dust Diseases Tribunal Act 1989 (NSW).
 Trowbridge Report, above n 32, vii–ix.
 I appreciate the vagueness of this term, but for me it captures an important truth about the enterprise of the Review. For a close analysis of ‘situation sense’, see William Twining, Karl Llewellyn and the Realist Movement (1973) 216–27.
 Trowbridge Report, above n 32, 27.
 ABC Radio National, ‘Public Liability’, Australia Talks Back, 3 September 2002.
 Spigelman, above n 41, 437–40.
 The judicial U-turn is documented in Harold Luntz, ‘Torts Turnaround Downunder’ (2001) 1 Oxford University Commonwealth Law Journal 95. Ironically, however, the liability-expanding decision in Brodie v Singleton Shire Council  HCA 29; (2002) 206 CLR 512 was one of the catalysts of the pressure for reform of public liability law. The judgments of the majority in the recent quantum-inflating decision in Cattanach v Melchior  HCA 38; (2003) 199 ALR 131 could be read as a determined attempt to distance the Court from the political arena of tort reform and reassert the supposed neutrality of ‘legal principle’. The superiority of ‘principle-driven’ to ‘underwriter-driven’ reform was influentially championed by Chief Justice Spigelman (above n 41, 440) and the idea of principles-based reform was taken up in the terms of reference of the Review (Review Report, above n 4, [1.27]–[1.28]). But it is unclear what work the concept does except as a term of approbation of changes to the law favoured on other grounds.
 Cf Rob Davis, ‘Basic Children’s Rights Threatened by “Tort Reform”’ (2002) 54 Plaintiff 4, 4: ‘the authors of the report exhibited a blind faith that they, using the imperfect tool of the political process, could do a better and more balanced job of delicately defining the law than the courts’.
 See, eg, Review Report, above n 4, [4.23].
 See, eg, ibid [11.23].
  HCA 35; (2002) 191 ALR 449 (‘Tame’).
 See, eg, Civil Liability Amendment (Personal Responsibility) Act 2002 (NSW).
 This latter concern, I believe, underlies popular complaints about awards of tort damages — especially large awards — to ‘undeserving claimants’: why should they be so generously treated when others with similar needs must rely on relatively meagre social security payments? Many people find any very large award difficult to understand because such awards seem to represent wealth far beyond what the average person could ever hope to accumulate. At the time of the Review, the Calandre Simpson case provided a focus for such concerns: Simpson v Diamond  NSWSC 1048 (Unreported, Whealy J, 21 November 2001); Diamond v Simpson [No 1]  Aust Torts Reports 61-695.
 See Don Dewees, David Duff and Michael Trebilcock, Exploring the Domain of Accident Law: Taking the Facts Seriously (1996).
 See Cane, Atiyah’s Accidents, above n 7, 337–9.
 National Committee of Inquiry, above n 20.
 Harold Luntz, ‘Proposals for a National Compensation Scheme’  Law Institute Journal 745.
  EWHC 326;  3 WLR 1038.
 Ibid 1053.
 See Cane, ‘Distributive Justice and Tort Law’, above n 6, 405–6.
 See Australian Health Ministers Advisory Council, Responding to the Medical Indemnity Crisis: An Integrated Reform Package (2002) [7.39]–[7.61].
 On 7 August 2003, Senator Helen Coonan issued a press release confirming that ‘[a] long-term care scheme ... is still definitely on the national agenda’. According to the announcement, 61 per cent of catastrophic injuries are the result of road accidents, 13 per cent of workplace accidents, 11 per cent of medical misadventure and 15 per cent fall under the ‘public liability’ umbrella: Helen Coonan, ‘Long-Term Care Scheme Still Firmly on the Agenda’ (Press Release, 7 August 2003) <http://assistant.treasurer.gov.au/atr/content/pressreleases/2003/079.asp> .
 In Britain in the early 1990s the British Medical Association was a strong advocate of no-fault until the National Health Service took over responsibility for meeting medical negligence claims, thus removing doctors’ incentive to support radical change. Once again, however, the system is perceived to be in crisis and no-fault alternatives are back on the agenda: Department of Health, United Kingdom, Making Amends: A Consultation Paper Setting Out Proposals for Reforming the Approach to Clinical Negligence in the NHS (2003).