The Health and Moral Hazards of No-Fault Insurance
PUBLIC POLICY SOURCES, NUMBER 2

by Peter Sheldon

Contents

Introduction: The Health and Moral Hazards of No-Fault Insurance
Executive Summary
Introduction
Section 1: What is No-Fault Insurance?
Section 2: The Insurance Corporation of British Columbia
Section 3: Alterations in the Insurance Product Recommended by the Automobile Insurance Review Team
Section 4: Why No-Fault Is Being Considered In British Columbia
Section 5: No-Fault Insurance in Other Jurisdictions
Section 6: The Empirical Evidence Related to the Impact of the Introduction of No-Fault Insurance
References

The siren call of "pure" no-fault automobile insurance echoes once more in Canada. This time its seductive strains are heard in British Columbia. The imminent introduction of no-fault insurance raises familiar questions. "Is no-fault automobile insurance better than a tort liability system in containing premium costs, producing fairer awards and reducing injuries?" The quick answer, based on empirical evidence of no-fault insurance and economic theory, is that no-fault does not deliver the benefits promised; it simply redistributes the costs which may actually increase.

This paper provides the somewhat longer answer to that question. The Fraser Institute, through its Law and Markets project, is pleased to publish the work of first time Institute author, Peter Sheldon, one of Vancouver's up and coming practicing economists. This introduction sets the stage for Sheldon's detailed analysis. In turn, his work will set the stage for later Institute investigations into the effect of no-fault insurance should it be introduced in British Columbia. This paper builds on the existing body of work done on automobile insurance in British Columbia, most notably, Herb Grubel's 1985 book for The Fraser Institute, Focus on the Insurance Corporation of British Columbia: Public Monopolies and the Public Interest.

First, here are some quick definitions. A tort liability system allows automobile injury victims full access to the courts to sue at-fault drivers and/or their insurance company for losses, economic or personal, beyond what is covered by the insurance of either or both parties. A pure no-fault insurance system does not allow injury victims to pursue compensation in the courts. Rather, all benefits are set by the insurer according to a predetermined scale. Variations of the no-fault regime include add-on insurance that sets a basic amount which injured parties receive whether or not they were at fault. Most provinces, including BC, have some form of add-on no-fault insurance. Other mixed no-fault schemes allow victims to sue if they meet either a "verbal threshold," based on the severity of injuries sustained, or a "monetary threshold," based on the cost of care.

The origin of the debate

The origin of the current no-fault debate in BC lies with the increasing pressure on the finances of ICBC, the public monopoly for automobile insurance. BC is a tough market for any insurer; it has the highest rates in Canada for both automobile accidents and car theft. The average cost of injury claims is estimated to have nearly doubled from $12,500 in 1985 to $21,404 in 1995. Stolen vehicle claims have gone from 4,876 in 1986 to 28,000 in 1996. In addition, the provincial government has off-loaded onto ICBC many of the costs of the Motor Vehicle Branch and of road safety initiatives, including the controversial photo radar. Not surprisingly, average ICBC premiums have risen by 155 percent since 1986. Last year, ICBC predicted that the current liability system would force up premiums by another 40 percent by the year 2000 if nothing is changed. On the eve of the 1996 provincial election, Premier Glen Clark promised that, if elected, premiums would not increase.

In December 1996, the recently elected Premier Clark then set up an Automobile Insurance Review Team to provide "options" to reduce ICBC costs while keeping average premiums capped at $975 through the year 2000. In March of this year, it proposed two possible insurance product changes. The first would modify the current liability system but retain a distinction between at-fault and innocent claimants and allow access to the courts. However, ICBC would get first crack at any pain and suffering awards through a $35,000 deductible, and it would manage any cost-of-care awards (victims would not control any lump sum award). The second option, recommended by a Canadian Bar Association committee headed by Thomas Allen, Q.C., would create a pure no-fault system in which an alternative dispute resolution process would replace access to the courts. ICBC would guarantee income replacement benefits of 100 percent of net income-rather than gross income as is currently the practice-to a maximum of $650 per week. However, ICBC would deduct any other CPP or Employment Insurance income, and would provide no compensation for pain and suffering, and only limited amounts for loss of function.

The government is expected to introduce legislation this Spring.

In preparing public opinion for this controversial step, the government has directed ICBC's media campaign. The message is surprisingly stark. Lawyers-and by extension the courts-are portrayed as impeding the efficient, equitable, and affordable resolution of automobile-related personal injuries. The claim is that if the formal legal process could be circumvented, then consumers would pay lower premiums, victims would receive fairer compensation, and ICBC could spend more on safety measures to reduce the number and severity of car accidents. In short, ICBC and the government claim no-fault insurance will bring considerable savings and safer roads.

The evidence on no-fault insurance

Does the empirical record support ICBC's contentions? This is the central question addressed by Sheldon in his work. New Zealand adopted the first pure no-fault scheme in 1974, followed by Australia's Northern Territory in 1979. In Canada, limited add-on no-fault insurance schemes have been adopted by several Canadian provinces starting with Saskatchewan in 1946. In 1978, Quebec was the first North American jurisdiction to move to a pure no-fault regime. Manitoba followed suit in 1994, and Saskatchewan in 1995. Ontario introduced no-fault insurance under a "verbal threshold" in 1990, graduated to a pure no-fault system in 1994, then returned to "verbal threshold" in 1996. Vancouver economist Peter Sheldon recently completed a survey of the various empirical studies about the effects of no-fault insurance and the results world-wide are consistent.Peter Sheldon, "No Fault vs. Tort Insurance Schemes: A Survey of the Empirical Evidence," Associated Economic Consultants Ltd., Vancouver, 1996note

One of the most controversial issues is the impact of no-fault insurance on the rate of accidents, injuries, and fatalities. Sheldon discusses two major empirical studies of Quebec which point out that since 1978, both fatal accidents and accidents leading to injuries have increased.Both studies are summarized in Peter Sheldon, "No Fault Insurance and the Debate in British Columbia," The Fraser Institute (Forthcoming, May 1997); M. Gaudry, "Measuring the effects of the No Fault 1978 Quebec Automobile Insurance Act with the DRAG Model," in G. Dionne (ed.) Contributions to Insurance Economics, Boston, MA: Kluwer Academic Publishers, 1992, pp. 471-498; and R.A. Devlin, "Liability versus No Fault Automobile Insurance Regimes: An Analysis of the Experience in Quebec," in G. Dionne (ed.), Ibid., pp. 499-520note Professor M. Gaudry found in his 1992 study that an estimated 6.8 percent increase in fatalities, 26.3 percent increase in injuries, and a 11 percent increase in property damage can be associated with Quebec's introduction of no-fault insurance. In a separate analysis, Professor R.A. Devlin found similar order of magnitude increases correlated to Quebec's no-fault scheme: 9.62 percent increase in fatalities, 27.0 percent increase in injuries, and 5.33 percent increase in property damage. Professor I.R. McEwin, in examining automobile injuries in New Zealand and Australia's Northern Territory following their introduction of pure no-fault insurance schemes, found the same effect-more deaths, by up to 16 percent more.I.R. McEwin, "No Fault and Road Accidents: Some Australasian Evidence," International Review of Law and Economics, vol. 9, 1989, pp. 13-24note

Detailed research has yet to give the full picture on how no-fault insurance affects the costs of premiums across Canada. In any event, conclusions would be hard to draw given the prevalence of public monopoly insurers who often set premium levels by political rather than actuarial calculations. Still, it is worth noting the "coincidence" that the provinces which have had pure no-fault insurance since 1993 have witnessed the highest premium increases: Quebec up 35 percent, Ontario up 29.5 percent, and Manitoba up 12.9 percent.

The major financial effect of no-fault insurance-and hence its attraction-is that it imposes a cap on individual injury awards and on their associated transactional costs, e.g., legal counsel and court fees. Yet, at least in the case of awards, this can be an illusory gain if the government is both the monopoly insurance provider and the monopoly health care provider, as is the case in British Columbia. If ICBC is able to secure a cap on individual awards, that may still leave unpaid bills for long-term care, for instance. The money has to come from somewhere, whether it's from the individual or from the public health care system. If the individual cannot pay (and it's likely he or she cannot because they are not working), then ICBC's "savings" shows up as an added expense in the health care budget. What British Columbians may save in lower insurance premiums they will have to make up in higher health care premiums.

For most economists, the potential problem with no-fault insurance is far greater than the unintended displacement of costs. No-fault insurance may actually encourage anti-social behaviour, despite all the good intentions of its designers. In terms of automobile insurance, the anti-social behaviour is, of course, reckless driving. Economists describe the effect as "moral hazard," and it can be particularly acute under the conditions of a public monopoly insurance provider.

A good way to illustrate "moral hazard" is by using the well-documented fact that restaurants insured against fire burn down more often than restaurants that aren't. The reason is not that insured restaurant owners tend to be immoral arsonists. Rather, it is that the owners, knowing that they have insurance, may not take due care to prevent fires. They may cut costs on sprinkler systems and on grease cleaning schedules.

Proponents of no-fault insurance, however, argue that moral hazard is not a factor in driving behaviour. They point out that driving is such a complex activity, requiring practically hundreds of driver actions per minute. They contend that "honest mistakes" rather than negligence account for most accidents; and, if we are all equally susceptible to accidents, it is unfair to discriminate against those who do get into accidents.

Can "moral hazard" exist in the relationship between driving and automobile insurance? Well, the empirical evidence from Quebec, New Zealand, and Australia does indicate a link.

Economic theory also suggests there should be a connection because price and demand are inversely related. That is, if you raise the cost of reckless driving, you will reduce the demand or the likelihood of bad driving. A driver will curtail the thrill of speeding or the convenience of driving home drunk if he or she knows the cost of an accident could prove catastrophic.

That some drivers are not constrained by price does not invalidate the existence of moral hazard. In theory, insurance companies adjust for the distribution of drivers on the risk-adverse-to-reckless axis in their pricing of coverage premiums. They modify how their clients drive by raising the costs of coverage to a point where it deters anti-social behaviour, yet is still affordable. A well-informed insurance market estimates the optimum balance between risk and premiums by compiling information both on the risk characteristics of its clients and on the likely range of awards paid out when accidents do occur.

This is where tort penalties for automobile injuries play a vital role in the efficient workings of the insurance market. Court awards for economic and personal losses provide a reliable guide to the true cost of an injury. The problem with awards set under a no-fault insurance regime is that they fail to reflect the complete cost of injuries both to individuals and to society.

This failure under no-fault insurance to generate complete information on costs can, in turn, lead insurers to under-price premiums which thus, in effect, subsidizes reckless drivers. This is the moral hazard effect.

In the case of BC, no-fault insurance may add to the existing risk of an induced moral hazard effect created by a monopoly insurance provider, ICBC. A monopoly insurance provider, mandated by law not to fully differentiate risk in setting premium prices for various client categories, already charges low risk clients too much, and high risk clients too little. (The specifics of the interaction between a public monopoly insurance provider and no-fault insurance regime will be investigated in later Institute research.)

However long the answer, both experience and theory suggests that no-fault insurance fails to meet the grade of equity and efficiency. Awards to victims are less than the true cost of their injuries. Hidden costs are passed on to either the least able-the victims themselves-or the least suspecting-the taxpayer. Bad drivers are subsidized at the expense of good drivers and, as a result, risky driving is rewarded. As reckless drivers are unchecked, accidents increase, and everyone ends up paying more.

As Sheldon has worked independently, his conclusions and work do not necessarily reflect the views of The Fraser Institute and its Directors. The Institute does, however, welcome his contribution and hopes that it will shed important light on this controversial topic. Public responses to this document are welcome and should be addressed to Owen Lippert at The Fraser Institute. -Owen Lippert

This paper examines whether or not auto insurance reform is appropriate for British Columbia. In recent months, BC's New Democratic Party (NDP) government and the Insurance Corporation of British Columbia (ICBC), the public provider of insurance, have repeatedly claimed that costs, particularly in terms of settling bodily injury claims, have been increasing at a rate over and above what one would anticipate given the growth in the number of drivers and the general level of inflation.

To substantiate their claims, ICBC contracted the consulting firm KPMG to conduct a number of studies on its behalf. These studies examined the cost history of ICBC, and set out a number of insurance product changes that would help alleviate the financial concerns of the corporation. The product reforms suggested by KPMG were all variations of no-fault insurance which were designed to restrict access to courts and to contain the level of premiums at current levels. Further, Mr. Doug Allen was appointed in late 1996 to head a commission that would look into auto insurance reform. The conclusions of this commission were twofold.

The first conclusion was that a range of measures should be introduced to help reduce the level of accidents. Second, in terms of product reform, two options from the KPMG analysis were put forward. One option amounted to a pure no fault scheme, with access to the court system denied. The second option retained access to the courts but imposed a heavy deductible on pain and suffering awards. Both options increase the benefits currently available to at-fault drivers.

This study notes a number of concerns with respect to the analysis conducted by KPMG and the Allen Commission. Of primary concern is that the work of KPMG received severe criticism by opponents of no-fault insurance who claimed that KPMG did not properly analyze the sources of the cost increases of ICBC, and further, that the cost projections provided by KPMG were based on data that was not replicable in the absence of data from ICBC. Also, a question exists as to the extent of the so-called "financial crisis" at ICBC. Moreover, a body of empirical evidence has been published which suggests that the introduction of no-fault schemes in other jurisdictions has increased the number of accidents and fatalities through moral hazard effects. This later point was not acknowledged by either the Allen Commission nor in any of the KPMG reports.

At the time of writing the NDP government is pondering whether or not to introduce a new insurance product in BC. A host of possibilities exist for reform. In this article's conclusion, additional cost saving measures are suggested, all of which reduce the insurance coverage available to accident victims. These measures include fining at-fault drivers a flat rate for each accident caused and imposing a significantly smaller deductible on pain and suffering awards. The calculations performed with limited data suggest that a deductible of $10,000 on pain and suffering awards would achieve significant savings for ICBC. Additionally, scope for additional savings is possible with the adoption of alternative dispute resolution procedures, along with a number of ways in which ICBC manages its own affairs. The latter include better management of investment funds and improved performance in collecting receivables.

As has been widely acknowledged, including by the Allen Commission, the only certain way to reduce insurance costs is by reducing the number of motor vehicle accidents. No-fault insurance schemes cannot by themselves contain the growth in insurance costs.