Facility Proposal Post-Mortem

The debate surrounding the future of Canadian residual markets for automobile insurance has been particularly heated in the last few months. Facility Association members in June 2007 declined to approve a proposal to enable the creation of Risk Sharing Pools (RSPs) as the primary residual market mechanism for all private passenger vehicles.

Despite myriad operational improvements that have been and continue to be made to existing mechanisms, the question of what, if anything, to do next from a governance and framework perspective is no less contentious than the debate that swirled around the failed proposal. And for good reason: any change to the residual market structure has the potential to affect all stakeholders in the automobile insurance arena.

As we ponder the future, perhaps now is a good time to revisit the history of residual market mechanisms, why they exist, what forms they can take and the pros and cons of each.

So far, a competitive market has proven to be the most efficient and effective mechanism devised to maximize the availability of goods and services to people at the lowest possible cost. But “most efficient” doesn’t mean “entirely efficient.” A market is only 100% efficient when all willing buyers are matched up with willing sellers.

In our business, a broker or agent usually brings willing sellers and willing buyers together. Some customers can be left over after this “mating ritual” is finished; by definition, this is the residual market. In jurisdictions in which auto insurance is mandatory, there has traditionally been a “market of last resort” to guarantee service to these customers. A notable exception is the United Kingdom, where there is no residual market mechanism.

A BRIEF HISTORY

The traditional Facility Association Residual Market (FARM) mechanism was established in 1979, modelled after Joint Underwriting Associations (JUA) in the United States. The FARM works using a small number of companies as servicing carriers that share financial results across the industry on a market share basis. This system has three key advantages:

* it provides universal access to coverage for the consumer at the distribution level;

* there is no impact on relationships between intermediaries and their companies; and

* it shares financial results equitably among companies regardless of their size.

The FARM does relatively well at coping with routine market inefficiencies. However, guaranteeing availability in the event of market failure, defined as a significant and relatively sudden decline in voluntary supply, is another story. In our business, large markets can fail for a number of reasons, including sudden changes in claims costs, strategic shifts by large carriers or well-intentioned but restrictive pricing regulation. Smaller markets can fail due to something as simple as a broker in a remote area going out of business. When markets fail, North American governments (reflecting society’s wishes) are unwilling to let thousands of consumers go without automobile insurance while the forces of supply and demand restore market equilibrium.

The most recent example of a mass-market failure in Canadian automobile insurance — and arguably the most extreme in many decades — occurred between 2001 and 2004. Within that timeframe, the number of private passenger vehicles written through Facility Association went from approximately 61,000 to a peak in late 2003 of just over 336,000. About 275,000 Canadians who had previously been insured in the regular market, and most of whom did not become bad drivers overnight, were confronted with residual market prices often two to three times the industry average.

Though technically the FARM “worked,” in that it guaranteed the availability of automobile insurance for all eligible applicants, regrettably it did so in a way that consumers and their political representatives did not accept. The perceived unfairness of premium increases for some consumers drew negative media attention to our industry and contributed to unwanted government intervention in the market.

BUMPS IN THE ROAD

In addition to the higher rates applied to more than a quarter of a million consumers, and the associated political and media backlash, the crisis of 2001-2004 highlighted key operational difficulties in the FARM model. They included the challenge of scaling up to accommodate a five-fold increase in business volume in less than 24 months, and the challenge posed to the actuarial paradigm by not only the sudden change in size of the FARM book of business, but by the composition of risks within it as well.

Against this backdrop, the Facility Association board of directors met in June 2005 to revisit the issue of the most appropriate structure for residual markets. In doing so, the board looked at two other types of residual market mechanisms generally used for automobile insurance – Assigned Risk Plans (ARPs) and RSPs – with the intention of proposing to its members a model for the future.

In an ARP, someone who cannot obtain insurance has his or her policy written at the plan’s rates. A central office assigns the person to a company, usually on the basis of a company’s market share (i.e. a company with 10% of the market will receive 10% of the applications.). Typically, all intermediaries will have the ability to bind coverage on behalf of the plan, ensuring availability at the distribution level. ARPs are currently in use in more than 40 U.S. states.

RSPs act as an industry-supported reinsurance mechanism. The mechanism is invisible to the consumer: policies are written on company paper at company rates and customers receive service in the normal way. Companies can then decide to keep the business on their own books or cede it to the pool. The design parameters for a RSP will usually include an expense allowance for the ceding company, a limit to the number of exposures allowed to be ceded, a percentage of risk retained by the ceding company, and a formula for sharing results. RSPs do not, in and of themselves, guarantee availability. A concurrent “take-all-comers” environment is therefore needed to ensure customer access at the distribution level.

Facility Association’s board decided against recommending an ARP, primarily because of the relatively large policy limits in Canada (compared to those of the United States) and the potential for those limits to “throw a grenade” into the financial statements of smaller companies.

RECOMMENDATIONS PUT ON THE TABLE

Broadly speaking, the board decided to recommend keeping the FARM for non-private passenger vehicles and for private passenger vehicles and drivers that are obvious “high risks” (drivers with criminal code convictions, vehicles modified for speed, etc.). The board also decided to recommend some kind of RSP as the primary residual market mechanism for private passenger vehicles. At the time, three out of the six provinces the Facility Association served were using RSPs created in reaction to industry needs. The board believed it would be worthwhile to take a proactive approach and create a more harmonized framework across Canada; at the same time, it cited invisibility to consumers, scalability and actuarial responsiveness as advantages over the FARM model.

A multi-stakeholder, ad hoc working group spent 18 months on consultation and design to develop a model that created a useful RSP mechanism while at the same time minimizing the risk of the RSP being used as a source of competitive advantage by member companies. In the end, the working group developed a model that called for separate pools in each jurisdiction for new vs. renewal business, each with its own distinct parameters for risk cession/retention percentages, transfer limits, and sharing formulas.

The renewal RSP was designed to be relatively “open,” enabling companies to retain business that had deteriorated on their books for the time required t o restore pricing adequacy, as opposed to having the company non-renew it into the market.

The new business RSP was designed with significantly more restrictive parameters, but flexible enough to allow companies to take on risks they would be required to accept.

THUMBS DOWN

Ultimately, the board viewed the working group model as administratively complex; it proposed a simplified version to members earlier this year. That simplified proposal failed to gain the necessary member support. A key focal point for opposition was a 5% transfer limit calculated at the group, as opposed to the company, level. Many voiced the concern that such a transfer limit could open the door for large companies to deliberately target and underprice market segments, forcing smaller companies out of the market at the expense of the industry as a whole.

Even had the proposal to amend the Facility Association Plan of Operation passed by a narrow margin, it would have been unlikely to receive the necessary regulatory approval to proceed. Regulators have suggested that any such reforms must generate a demonstrable improvement for the consuming public over the current framework and have the support of a vast majority of insurers and intermediaries.

It would be too simplistic to say there were two camps in this debate — i.e. those in favour of the simplified proposal and those in favour of the more complex one. Differing views were expressed regarding whether cession limits should be at the company or group level, whether ceding companies should retain a percentage of risk, and on other issues as well. Some companies that voted against the proposed model thought it did not adequately address the risk of predatory pricing. Others were simply opposed to any kind of RSP, and wanted to see the problem addressed by working within the FARM framework.

A WORK IN PROGRESS

The voting results serve as a reminder that our industry is far from the monolith that many portray it to be. Yays and Nays included both large and small companies, foreign-owned and domestic, as well as companies with direct and broker distribution methods.

Looking ahead, we can expect any further proposals for residual market reform to be hotly debated: they certainly have been in the past, both here and throughout the United States. Past proposals have not achieved consensus to this point, but may contain the building blocks that, combined with new ideas, will result in a workable model for all. As we consider options for residual market reform, we must remind ourselves “the devil is in the details.” We must carefully weigh and balance the benefits and risks of likely outcomes against potentially adverse impacts on industry stakeholders (both company and distribution alike). We also need to remind ourselves that any proposed framework must support the competitive market to the ultimate benefit of all Canadian auto insurance consumers.