Nathan Skinner chairs a discussion on risk management in insurance companies
Let’s start by looking at Solvency II, a major piece of legislation for the insurance industry. What are the challenges and opportunities?
Oliver Peterken: I would say the modelling is both a strength and a weakness of Solvency II. It’s a strength because, for the first time, it starts to simulate the risks a firm is taking onto its balance sheet. I think everyone accepts it is conceptually a good step forward. The weakness of course is the actually ability to capture the complexity of an insurance operation. There’s probably a lack of experience among all the regulators in understanding the weaknesses of modelling. Whereas in practice the people who are actually building the models know how complicated they are and how many assumptions have to be made and how variable the results can be.
A major feature that has been overlooked is the increased cost and complexity that it puts on smaller organisations. If you look at some of the costs of complying, they will have a substantial impact on smaller insurers. Combine that with what it will probably do to capital requirements, and you could see a lot of capacity leaving the market on certain risks.
Richard Winter: The impact on small companies is far more than just the financial impact. It’s going to be a drain on management’s time. There’s a big difference between a model that gives you insight into the business and one that is approved for Solvency II. I think that most companies, certainly in this country, would have the basic models and be learning things about their business. But taking that and going through the Solvency II hoops is an overhead that may add very little value. If the capital is available they may as well just take a slightly bigger hit and get on with running the business.
What do you think about Solvency II and the current economic situation? Are there political considerations?
Oliver Peterken: It is unfortunate timing for the supporters of Solvency II that they are trying to get it through the European Parliament in the middle of probably the worst financial crisis the world has ever seen. With the background of various concerns over regulatory failures in numerous jurisdictions, it is going to be a much more difficult political process than it would have been six months ago.
Six or nine months ago Solvency II would probably have been nodded through the European parliament and most national parliaments. I think now it will be caught up in all the questioning and the various political investigations into what’s gone wrong. And you can’t see it would necessarily get through that process completely unscathed. I think the political environment has shifted fundamentally as far as anything to do with regulation is concerned.
I’ve never known a measure that has had such widespread support as Solvency II. I think the delays have served to get people behind it. But it was conceived over the last ten years or so in a very different economic and political climate to what we are now seeing. I think the political factor has become quite a big unknown.
Richard Winter: I think the timing issue is interesting. It would be very hard to stop the process altogether. But to let it time out and let it drift on a few years, and then come back again may well happen. I think it’s going to quickly become clear whether it’s still going to the same timetable, in which case there’s a lot of work to do very quickly.
Let’s assume it does go ahead
Oliver Peterken: There are some very real issues if we assume it goes through largely unchanged. If you look at the number of insurance companies in Europe and the actuarial and risk management resources that are available to get models approved, it’s physically not there. You have a scenario where you could have regulators, consultancies and insurers bidding up the costs of the resources for doing the job. I question the extent to which the supporters of Solvency II thought about how you actually roll this out. In the UK we are in a privileged position. In other parts of Europe you don’t have that level of expertise.
David Innes: And the level of understanding of models isn’t necessarily there. As well as the time needed for companies to develop their approach to modelling, the regulator has to have the resources to understand the approach taken and the model used by each company. As every company is different a good model for one company would not necessarily be a good model or approach for another. In many ways the least important part of solvency modelling is the identification of the actual amount of capital required, more important is the process around and the identification of the real risks.
Oliver Peterken: One of the great successes of the EU has been the reinsurance directive and opening up the single market. There’s undoubtedly regulatory arbitrage going on. You can already see some of the major European insurers changing their location. If you get different implementations of Solvency II you will get companies migrating to the least onerous jurisdiction. If you couple that with tax, you’re going to start creating some very uneven playing fields.
Richard Winter: But isn’t SII going to help that rather than hinder it? At the moment there are very different standards. At least with Solvency II you have a common purpose. They may be doing it in different ways. But it will be closer to a level playing field than what we have now.
Oliver Peterken: I agree with you that it will level it out compared to now, but I don’t think it will be as level as they were intending.
David Innes: You have the interpretation issue as well. Some countries are very prescriptive while others are laissez faire.
Shall we move onto ERM? Could or would the ERM approach have prevented some of the problems we have seen recently?
Oliver Peterken: I think the supporters of ERM do like to offer it as this wonderful saviour. That, somehow, if only companies would implement it properly it would make everyone risk free. That argument is usually made by consultancies who want to sell you rather expensive large scale ERM projects.
It’s important in the middle of a financial crisis to go back to real evidence and cut through the noise. In the middle of a crisis, all you can see is the symptoms and the pain around you, understanding is very difficult. What most companies do is just focus on managing through it.
The Swiss regulator has published an edited version of the report that UBS filed after it lost $37bn earlier this year. It was UBS’s own internal audit report about what went wrong. It wasn’t a question that ERM wasn’t implemented. UBS seemed to have had some quite good systems. It was that some absolutely fundamental risk management things didn’t happen.
For example the trading desks which got into problems didn’t have any limits on how much of the credit derivative products they could hold at any one time. They were effectively able to have an unlimited warehouse of risks. Then there was senior management that was very keen to get into this area, feeling they were being left behind. You had the UBS risk management function not looking at this area, because it was only about 5% of the total risk capital and it wasn’t considered to pose a major threat to the business.
Now you don’t need ERM to tell you those problems; you just need some very basic risk management. So I’m not sure ERM is the panacea for current issues. I suspect that UBS will not be alone in having ignored some fairly basic things.
David Innes: If you step back and look at the housing market, it’s very similar to what happened in the UK in the early 1990s. The basic lending parameters are very similar. Relaxation of lending quality has led to an explosion in credit. What we tend to do if we are not careful is forget what happened in the past. Or we think the risks are mitigated and those things can’t happen again.
Oliver Peterken: In the last 10 to 15 years, the prevailing political climate was very critical of any intervention in the operation of markets. If any regulator 10 or 15 years ago had said we are very concerned about the credit derivatives market and we think it should be brought under our regulatory jurisdiction, you can imagine the reaction from the banking sector. They would have shouted them down.
Richard Winter: A company with good ERM is more likely to avoid problems than one that hasn’t got that structure in place. You can have the most fantastic set up in the world but if you don’t identify where the risk is coming from properly then you’re not managing it. At the moment there isn’t one thing that happened, there’s a whole series of events across a load of different areas. It’s that combination which is really hard to identify. The key thing with ERM is to try to push it on. If you ever think as a company that you’ve done it then you are doomed. You may have a very good process in place, but you need to keep feeding things into it and exploring more what can go wrong, and how you can control it better.
Oliver Peterken: Regulators say you should look at scenarios. There is no numerical model in the world that could have picked this up. If you had said two years ago here is a scenario we should test our business against. It is that the largest insurance company in the world is going to be nationalised by the US government and default on all of its financial products, and Wall Street is effectively going to disappear within the space of a fortnight. They would have said you are barking mad. They certainly wouldn’t have encouraged you let alone support you to test your company’s resilience against that scenario.
David Innes: Twenty years ago there was limited insurer exposure to financial products. One notable exception being mortgage indemnity, and we know what happened in the early 90’s to the UK property market and the consequences to mortgage indemnity insurers. As financial products and the corresponding insurance wrapping has become more innovative – we are left with things getting ever more complicated and the associated risks moving around. In the end, are we sure that we really know who is left with the risk?
What are the hallmarks of good ERM?
Oliver Peterken: If you’re not getting the right culture and tone at the top of the organisation it’s very difficult. You have to start with the CEO, not just supporting, but being proactive about the company’s attitude to risk and risk management, otherwise you don’t get a consistent message. You also have to have a clear and consistent approach to dealing with serious problems. Do you go down the no blame near miss reporting, or the 100% compliance with everything? You can implement ERM under both regimes, but you need to be clear.
Richard Winter: People at the top have to set the agenda. And you have to keep banging on about embedding it, until it becomes ingrained as one of the things that people think about.
In terms of getting that buy-in it’s partly about selling the benefits to individuals. It’s also about trying to get people to see how their job contributes and to think in the same structures as the top level of the company. Once you’ve got that you have the risk management culture embedded completely and it will really help peoples’ work. We’re a few years away from that.
Nathan Skinner, associate editor, StrategicRISK, chaired the debate. Participating were: Oliver Peterken, chief risk officer, Aspen Re; Richard Winter, company actuary, inter Hannover, and David Innes, chief actuary, RSA