Q&A: Osfi chief champions Canada's regulatory framework

Risk: Canadian banks have withstood the severity of the financial crisis better than banks in many other countries - what do you attribute that to? How much has the regulatory framework contributed?

Julie Dickson: There were a lot of factors in play so you can't really point to any one reason, but there were a few things that contributed. Firstly, we had a very different mortgage market in Canada so all high loan-to-value ratio loans had to be insured, typically by a government agency or an agency with a government guarantee.

Secondly, there were things Osfi did that proved to be important, mainly on the capital side in that we paid great attention to the quality of tier-one capital. Prior to the crisis we stipulated that 75% of tier-one capital had to be in common shares, and we set targets of 7% tier-one and 10% total capital; most banks tended to be well above that when the crisis started. We really focused on banks having robust plans in place to determine what level of capital makes sense for their risks.

Thirdly, Osfi's mandate is quite unique compared with the mandate of many other supervisors, in that it was totally focused on solvency, meaning that our role is to understand the financial condition of a bank and then to take action expeditiously to deal with problems as they arise. The mandate never referred to things that I've seen with some other regulators such as the need to take into account the competitiveness or innovativeness of the financial sector. That was very important in terms of the focus that Osfi brought to bank supervision. If the mandate is that you should take into account the competitive position of the country in which you're operating, you're probably going to take different decisions than if the mandate says you should understand the financial condition of a bank and take action early to try to deal with problems.

The other aspect of this question is the way in which regulatory agencies did on-site supervision. There is a lot of discussion at the moment about capital rules and leverage rules, but as important is how you go about the day-to-day supervision of financial institutions. Some regulators outsource that supervision, but we do it all in-house and we think that's very important. We employ just over 500 people here, but a lot of our supervision is reliance-based, meaning we place a lot of reliance on senior management and boards of financial institutions, but we test whether we can rely on them. If you compare us with the US, with its very different model of supervision, they have far more supervisors per institution than we do.

Risk: Has the financial crisis brought to light any particular weaknesses within the Canadian banking sector? Has it necessitated any changes to your regulatory approach?

JD: There are changes in the works to the way in which we regulate and supervise. I don't think that reflects a particular weakness in the banking sector, but it certainly reflects things we're learning as a result of the crisis. We did not spend a long time looking at compensation practices and it's been agreed that is something that is quite important and can have an impact on risk, so we're picking that up now. We would also agree that the framework for assessing liquidity needs to come up a notch, so we're working through the Basel Committee on that.

One other area we're focusing on is assessments of boards of directors, looking at both composition and effectiveness. In terms of composition, there needs to be more people on boards who understand and have experience of managing risk. The lack of risk professionals on boards is not unique to Canada; when you look at boards of financial institutions around the world, they often tend to be made up of people who have been in particular industries for a long period of time, but definitely not on the risk management side. If there were more risk-minded people on bank boards, there would be better oversight of management and deeper discussions. Some banks that have already done this have commented to us that it has resulted in much richer discussions at the board about the risks the bank is taking.

Risk: There has been strong criticism of Basel II over the last year - that it failed to set effective regulatory capital levels for banks and is excessively pro-cyclical - do you think that is fair? Which changes to the accord do you believe to be most urgent?

JD: I think the criticism of Basel II is overdone, because it wasn't in effect in the US before the financial crisis and has still not been implemented. In Canada, we implemented Basel II in 2007. A big part of implementing the accord is to improve risk management systems and to have better data so that you can assess your risk. That is a very positive aspect of Basel II and the chief risk officers I talk to here will admit that of their own accord. That said, a lot of people have talked about Basel II being excessively pro-cyclical, and, for countries that have introduced it, it is causing capital requirements to spike in the recession. Our experience would suggest that banks that did a good job on developing through-the-cycle estimates of probability of default and other measures are not seeing their capital requirements spike. But for those banks that have more of a point-in-time methodology, you certainly do see more pro-cyclicality. Any risk-based system is going to have an element of pro-cyclicality, but how much is acceptable is something we really do need to keep discussing at an international level.

Risk: Would you support the introduction of dynamic provisioning, so that banks are mandated to build up capital buffers during the good times that can be drawn on during a recession?

JD: As a bank regulator, I would like to see more flexibility in the accounting rules to allow loan loss provisions to be built up. In our experience in Canada, it was difficult for banks to build up those provisions even though they might have felt that risk was building in the system. As a bank, you can increase your capital levels if you feel risk is increasing, but you're far more constrained on loan loss provisioning given the way that the accounting rules work. So I do think more needs to be done there. There are lots of discussions going on and the International Accounting Standards Board (IASB) has indicated its willingness to discuss it. I look forward to those discussions continuing between the Basel Committee and the IASB because I think it's a very important issue.

Risk: The Basel Committee is also looking at the possibility of introducing a gross leverage ratio for banks, something Canada already has in place with its maximum assets-to-capital multiple. Do you think rolling this measure out internationally would be an effective way of constraining bank leverage in future years?

JD: We have had the leverage ratio for a long time and we think it's an important tool. It's very difficult to measure risks with 100% accuracy, so having a leverage ratio on top of risk-based capital is an added measure that helps us sleep at night. We're in discussions at the Basel Committee on a leverage ratio to be used internationally. There are many ways to design it, so that's an important element of the discussions.

See also: Raiders of the loss arc
Crude but credible