...So as we work on the international capital standard for insurance, as we move through the next round of ORSAs, and as we consider the prudential implications of new uses of reinsurance, we will continue to consult the industry, and keep the dialogue going.
I am pleased to be here to speak with you this evening about some of the issues currently facing the property and casualty insurance industry in Canada.
Many of these issues have their origin in technological change both past and future. Cyber security issues create, at the same time, both a threat to individual P&C insurers and a potential new business line. The growing ability to collect detailed data about policyholders and their behavior creates new options for pricing risk and new ways to incent policyholders to reduce risk. Distribution channels are evolving as new technology is developed and adopted and as customers get used to using new channels.
Turning to natural disasters, we know that severe weather events have been on the rise, and we have to presume that it will not get any easier to predict both the occurrence of these events and how they will translate into claims. And the spectre of earthquakes looms over eastern Ontario and the Saint Lawrence Valley in Quebec, and in the B.C. lower mainland and Vancouver island.
Most of you know these issues as well, in not far better than I do.
So, I will devote the rest of my remarks to regulatory and supervisory issues that are relevant to the P&C industry.
I will highlight three topics:
- The ongoing effort to create an international standard for minimum capital requirements for internationally active insurance companies;
- The implementation of the Own Risk and Solvency Assessment initiative in the Canadian P&C industry; and
- The supervisory implications of the growing use of unregistered reinsurance by some parts of the Canadian P&C industry.
International Capital Standards for Insurance
Let’s begin with the international insurance capital standards currently being developed at the IAIS.
We at OSFI strongly support this project. We look for two potential benefits for Canada.
First, it would be very useful to have a common yardstick for comparing insurance capital requirements across jurisdictions. This would better position us to assess any concerns that our capital standards were unjustifiably impeding the competitiveness of Canadian firms. Moreover, there is probably useful information in any major differences across countries in capital requirements that this exercise may turn up. It is too much to expect that the international framework will allow us to make very accurate “apples-to-apples” comparisons of our capital standards to others, at least in its first iteration. But it promises to be an important step forward.
The second potential benefit is to promote a level playing field for internationally active insurers. International agreements on minimum standards help all countries to impose appropriate prudential standards without having to be concerned about unwittingly degrading the competitiveness of their own firms.
We have been quite active in the development of the proposed standard. While we are pursuing a standard that will work for all countries, we are also focused on getting to one that will work for Canada. To that end, we have consistently promoted the need for flexible implementation that considers each jurisdiction’s legal and regulatory frameworks
Many of the pieces that are currently being discussed at the international level are already in place in Canada. So the work we have already accomplished positions our industry well for the eventual introduction of an international insurance standard.
I imagine that you would like to know what the impact will be on the Canadian P&C industry if and when OSFI adopts the new international capital standard. And I would like to tell you. But I can’t. The work on the new standard is far from complete, and calibration exercises have not yet begun. However, we currently believe that the international discussions are going in the right direction.
What I can do is to sketch out some possible scenarios regarding Canadian implementation. If we at OSFI find that the eventual international capital standard for insurance companies is too low for Canadian purposes, we will not relax our own domestic requirements. My predecessors did not race to the bottom when the first agreement on bank capital was completed, and all Canadians have benefitted from those decisions.
Moreover, if we feel that the international standard is not suited to Canada’s needs for some other reason, then we will simply continue with our existing approach. It is our responsibility to set what we see as prudent and workable regulations for Canada, and that responsibility will not be extinguished if there is an international standard for insurance capital.
Own Risk and Solvency Assessments
Let’s turn now to the introduction of the concept of ORSA in Canada.
Through the ORSA process, each insurer is expected to come to its own assessment of its risks, its capital needs, and its solvency position. ORSA is a tool to set out the links between an insurer’s risk profile and its capital needs, and so come to its own internal target for capital.
A robust ORSA is one where the company has comprehensively assessed the potential impacts of the risks it faces, and uses that assessment to make prudent decisions.
I cannot emphasize enough that the key word in “Own Solvency and Risk Assessment” is “Own”.
We expect all insurers to conduct a robust ORSA. Part of that expectation is for each company to apply the process in a way that is tailored to its own business. Otherwise, the ORSA process will not be effective, and the effort will be wasted.
I understand that many companies are interested in our feedback on the assessments that they have shared with us. Having seen the broad range of ORSAs, we at OSFI are well positioned to provide you with recommendations for improvement, and to compare your practices with those we see as the best in the industry.
And while we are well positioned to do all of that, we are not going to.
Why not? Because our overriding goal is to keep the “Own” in ORSA. If we started making specific suggestions to individual companies, we would, inadvertently, start substituting our judgment for yours. Of course we could try to reduce the impact of our comments by noting that they are only suggestions, that you are not obliged to follow our recommendations, that your ideas might well be better than ours, and so on. But I don’t think insurers ever forget, even for a moment, that we are the supervisor. And so comments that would be perceived as suggestions if they were made by anyone else come across as requirements if we make them.
What I can do is to share with you some broad observations about the full set of ORSAs prepared by the companies we supervise.
The ORSA reports from P&C companies generally provide a comprehensive overview of the insurer’s risk management framework. Nearly three-quarters of insurers outlined their risk appetite and linked it to risk. In many cases, risk tolerance and limits were also included and linked within the reports. This is a sound approach.
The reports we have seen generally provide a comprehensive list of risks, some with more than a dozen categories. Some insurers have preferred to segment risks, for example, treating reserving, pricing and catastrophe risks separately rather than bundling them as insurance risk. Others have taken the opposite approach.
We have not set standard definitions for risk categories to be used by the industry. That is deliberate on our part. Each insurer has the prerogative to build and define risks in ways that makes the most sense to them.
ORSAs are in their first year of implementation and there will be room for continuous improvement. Some elements and processes will need to mature before becoming all they can be. For example, some ORSA reports were largely descriptive. These had some value, but they were not necessarily useful for a Board or Chief Agent in understanding the risk profile of the institution. For a few insurers, the ORSA was not used to link the company’s risk assessment to its internal target for capital, as we would have expected.
We’ll have more to say about ORSA as we all get more familiar with the process.
Trends in the use of Reinsurance
Our third supervisory topic arises from recent developments in the use of reinsurance.
P&C insurers have long used reinsurance as an important risk management tool. It can be used to reduce insurance risks, to temper the volatility of financial results, to stabilize solvency, to make more efficient use of capital, to better withstand catastrophic events, to increase underwriting capacity, and to draw on reinsurers’ expertise.
Recently we have noticed that some firms have been shifting toward a business model of insuring commercial risks in Canada and reinsuring a significant portion, or virtually all, of the risk offshore. Often, risks are ceded to unregistered affiliates, with little capital retention in Canada.
This practice has long been used to transfer large catastrophic risks when there is limited reinsurance capacity in Canada. It has also been used by global insurers to service some of their large customers who have risks around the world, including risks in Canada.
What we are seeing more recently seems to fall into a different category. It appears to be being used by insurers to increase policy limits and sizes without changing net risk retention. This means that exposures ceded are quite large. It also means that exposures are going up without a commensurate increase in the capital of the direct writer.
Taken to extremes, this shifting business model introduces a very concentrated credit risk to policyholders. This raises prudential concerns given the possibility of distress in the unregistered reinsurer, whether affiliated or not.
History has shown that over-reliance on reinsurance can be dangerous if it is used as a means to “rent” capital to support rapid growth in insurance premiums. Reinsurance cover can evaporate when an insurer encounters stress, and this is usually happens when the company needs the reinsurance the most. This can leave the company in a struggle to renew its commitments. Ultimately, the company may be forced to quickly raise fresh capital to replace the disappearing reinsurance cover. And if it cannot raise that capital, it may fail.
Our Guideline B-3 provides valuable considerations on reinsurance risk management practices such as the importance of diversification of reinsurers and robust, frequent credit evaluation of reinsurers. Risk management practices should also be applied to reinsurance provided by affiliates.
We are currently assessing the scope and potential impacts that recent changes in reinsurance practices may have on the ability of insurance companies to meet their obligations to policyholders if the insurer were to encounter stress.
We want to ensure that when the use of reinsurance serves as a substitute for capital, it is not used in a manner that leaves policyholders less protected.
Having completed my tour of current regulatory and supervisory issues let me leave you with this.
For our regulatory and supervisory work to be effective, we need continuing input from the industry. We need information about developments in your businesses, we need your perspectives and ideas, and we need to know your concerns. So as we work on the international capital standard for insurance, as we move through the next round of ORSAs, and as we consider the prudential implications of new uses of reinsurance, we will continue to consult the industry, and keep the dialogue going.
I look forward to continuing our work with you as we face whatever challenges are ahead of us.