10 June 2009

Published in: Risk governance

Joachim Oechslin puts his faith in models at Munich Re

Quantitative modelling has long been part of the fabric of Munich Re. It's also a strength of the CRO but he doesn't believe in slavishly following the models, explains Sarfraz Thind.

Joachim Oechslin believes that a little pain is sometimes necessary to get you where you want. As chief risk officer (CRO) of the world's largest reinsurer, Munich Re, Oechslin feels that the company has learnt much from the mistakes of the past. "Munich Re suffered very much in the last (dot com) crisis," he says. "But we took very clear lessons from this. One of the major things we did was to reduce our dependency on the capital markets and make underwriting our core business. The current crisis is a capital markets crisis, but our focus has allowed us to come into this in an overcapitalised manner."

Certainly Oechslin's prior experience has helped him in his current role. Oechslin joined Munich Re in 2007 aged only 37. Prior to this he was deputy CRO at Axa, and before that he was at Winterthur, where he developed a strong reputation in the field of financial modelling. His paper on replicating financial options, published in 2007, was amongst the first to promote the use of replicating portfolios in the context of risk models. Since then portfolio replication has become a broadly used technique among insurers.

The focus on financial modelling is something that Oechslin has continued to emphasize in his current role. Since joining Munich Re Oechslin has boosted his quantitative risk modelling team from to 13 from nine, and he believes that the company currently has one of the best quantitative modelling set- ups in the insurance business. "Models are very important here-they are a direct input into the risk pricing that we charge. The modelling capabilities of Munich Re were established well before I joined, but we have been shaping them even more since I joined to improve them."

Working with RMS

The collaboration with US risk modelling firm RMS late last year should benefit the company even more, says Oechslin. The objective of this meeting of minds is to promote quantitative views on risk through use of scientific methodologies. The companies are working together to research the hazards and vulnerability of models as well as to understand underlying assumptions, uncertainties and limitations of quantitative modelling techniques. Oechslin believes that the work with RMS should enable Munich Re to boost its own modelling capabilities.

While the pre-Munich Re years gave him a good quantitative base, Oechslin's experience at his old employer Axa also benefited him in other ways. He is, for example, effusive in his praise of Axa's risk management structure. "Axa has a stringent risk management set up-it is a role model in some ways," he says. "You have the separation of group and local risk management which is something we are following too. For example we do not have total centralization but seek to combine both central and decentralized elements. This ensures that we have a top-down method through reporting lines but also allows the risk staff to be part of their individual risk management units. It also means that there is less ‘dotted-line' work done by me."

Risk management structure

The reinsurer currently has around 350 people involved in risk management worldwide with 60 to 70 organized into a central team led by Oechslin, and the rest reporting to the management of their individual business segments. These include Munich Re's primary insurance subsidiary Ergo which has its own CRO who is accountable to the Ergo board and also to Oechslin. Oechslin himself reports to the company's chief financial officer (CFO), Joerg Schneider, who is a member of the board of management. And he believes that this direct contact with a management board member is crucial to the reporting structure and one which serves the CRO role well. "This is an important aspect. I am not saying that the CRO needs to be on the highest management committee but if he is not then he needs to communicate his thoughts clearly to someone who is."

Munich Re also has a group risk committee which meets on a quarterly basis and is made up of the CFO, CRO and the risk managers of the various subsidiaries. On top of this, it has two separate risk committees for its individual reinsurance and primary risk lines to better manage the individual risks of the two divisions.

"We are now building up our credit risk exposure as we think that the spreads allow this. It is an anti-cyclical opportunity we now have."  

The time at Axa not only highlighted key benefits for his current role, it also taught him the differences between the risk management functions at a life insurance company and a reinsurer. Having seen both sides he believes that the role of a chief risk officer at a reinsurer is much more expansive than at a life insurance company, with the former more involved in assessing day-to-day strategic decisions on the business. "Being a reinsurance business our risk management approach is very different to that of the life insurer," he says. "In a reinsurance company the CRO is much more involved in taking risk decisions. In the life insurance line, the role is more about making policy decisions."

Accordingly, a large part of Oechslin's job involves analysing the risk behind potential merger and acquisition deals and assessing transactional risk on individual reinsurance deals. "In reinsurance on top of policy decisions you have large individual transactions where you need to judge the risk. For example, I am often involved in assessing the credit risk on a particular deal proposal which I may not do at a life insurer." And he is certainly being kept very busy analysing new opportunities right now.

Low exposure to credit

Throughout the bull market years of 2003 to 2008 Munich Re kept a fairly low exposure to structured credit compared to most of its reinsurance peers. While most who had exposure then are probably regretting it now, Munich Re is looking to go against the grain and increase its credit exposure. "We maintained a low exposure to structured credit even in the good years, as the company felt there wasn't enough compensation for credit risks," he says. "However we are now building up our credit risk exposure as we think that the spreads allow this. It is an anti-cyclical opportunity we now have."

Oechslin says that sifting out the best-yielding opportunities with the lowest systemic risk is likely to be one of the main challenges for companies like Munich Re right now. This positive note is quite different from many other insurers who are still trying to avoid the worst ravages of the credit crisis on their existing investments. To date Munich Re has made one major acquisition-the US primary insurance company Hartford Steam Boiler from AIG in December last year. It also concluded a quota share treaty in the life and health reinsurance segment which was disclosed in April, through which it expects to generate an additional €2bn annual premium income this year, says Oechslin.

Catastrophe risk

The other major focus for the company at the moment is natural catastrophe risk. According to figures which Munich Re drew up earlier in the year, 2008 was the third costliest year for losses from natural event catastrophes. The large number of tropical cyclones and the earthquake in Sichuan cost the industry a total of $45bn in insured losses. And Oechslin is clear about the new challenges the company faces. "At year-end 2008 we can see that capital market risks are not the dominant risk," he says. "Now it is truly reinsurance risk. In other words, storm Europe or an Atlantic hurricane."

The company has also sought to address possible impact of a global pandemic, through rigorous scenario analysis and the launch of a pandemic bond programme called ‘Nathan' early last year which transfers extreme mortality risks to the markets. However, Oechslin believes that the development of new medicines and increased co-operation between health organizations should better prevent pandemics, like the recent swine flu scare, from developing in the future.

To date, Munich Re's risk models have held up very well in dealing with catastrophe risks. The company's actual exposure to storm-related losses was very closely matched by the expected exposure models between 2006 and 2008. Part of this success is down to a major overhaul of its risk models in 2005 during the fateful hurricane season of that year. And the company is looking to implement further tweaks to capture the fall-out from the credit crisis on its risk models. In particular the pro-cyclical behaviour of credit risk models displayed in the recent credit crisis, and the calibration of spread volatility risk are currently being reviewed.

"At year-end 2008 we can see that capital market risks are not the dominant risk. Now it is truly reinsurance risk. In other words, storm Europe or an Atlantic hurricane."  

But while he has faith in Munich Re's models, Oechslin is critical of a blind reliance on them, as has been shown by many insurers and banks in the current financial crisis. "There were a number of risk management failures in the banking industry but this is not something that the insurance business was decoupled from," he says. "In part this is because we have been using similar models which, like the banks', have pro-cyclical elements. Many investors focused far too much on short-term return on equity over the past few years which led to pressure on banks to create models calibrated to historically benign periods."

In the light of this, the equity dampener clause included in the final version of the Solvency II text, agreed earlier in the year, is something that has irked him. "You shouldn't be looking to under-calibrate models to get nice returns," he says. "But that is precisely what I see with the equities dampener clause. Politicians will follow their national agenda but this (equity dampener) argument is clearly not sustainable."

Still, he is appreciative that there is an agreement now on the table. "People have been fighting about group support and equity issues-these are important issues but not important enough to let Solvency II go down the drain. That would have been completely wrong." He says that Munich Re is well ahead of Solvency II in most respects, especially in its modelling capabilities, but that there remains much work to be done for the industry in general. "Solvency II provides the framework but it needs further work in the implementation measures. There is a lot we need to learn from the crisis. For example we need to look at the lessons in terms of modelling credit risk when considering the implementation measures."

Back to top