This page contains a Flash digital edition of a book.
Interview: Adam Podlaha, Aon Benfield Analytics


Strength in Numbers


Multi-modelling and bespoke solutions are key to effective risk management, says Adam Podlaha, head of Impact Forecasting International at Aon Benfield Analytics.


models playing in risk manage- ment in Europe and how is Aon Benfield Analytics involved?


Q What role are new catastrophe A There are three aspects to this. First, models help insurers decide how much


reinsurance coverage they need to buy—cou- pled with other factors such as risk appetite. Second, they help to track the evolution of in- surers’ exposures, which can be valuable infor- mation for risk managers insuring new loca- tions.


Finally, certain parts of the models can also be used to influence primary underwriting by assisting insurers to understand the vulnerabil- ity of individual insured properties. We are also helping to add clarity to our clients’ portfolios by improving the quality of their exposure data and provide guidance on how modelling can be used for Solvency II purposes.


Q What further innovation in model foresee?


development in Europe do you A The industry is heading towards a multi- model approach, so creation of more


transparent models, even for territories and perils that are currently well modelled such as European windstorm, is clearly a need. Having just one model is dangerous, but three or four


Adam Podlaha, head of Impact Forecasting International at Aon Benfield Analytics


ing solutions offer increased transparency and customisation options. The model needs to be understood both by the modellers and the cli- ent in order to make the decision about how much reinsurance to buy. Our models are open and customisable on many levels so the client has a much better grip


“The model needs to be understood both by the modellers and the client in order to make the decision about how much reinsurance to buy.”


allow improved insight and enable clients to make better decisions on their reinsurance pur- chases. This aligns with a point made by a prominent scientist, who said it is better not to look for a precisely correct answer but to take account of the whole range of plausible an- swers and expect the correct one to be in that range, and is one of the reasons Impact Fore- casting is creating its own European Wind- storm model, which will be released next year. In addition to a new model Impact Forecast-


on what is happening inside the model. For ex- ample, a client informed us that an engineer- ing company could provide interesting data for our Austrian flood model in respect of the planned flood defences level. We were then able to alter one of the model’s components to give that particular client an insight into what effect these changes will have on the loss estimates when they come into force in a few years’ time. Impact Forecasting is also focusing on non- modelled perils, such as tsunami, and is ex-


4 | INTELLIGENT INSURER —BADEN-BADEN TODAY | Tuesday October 25 2011


panding in areas including life modelling for earthquakes and terrorism.


Q What role will your academic A Aon Benfield Re search is the collective


partners play in this innovation? name for the firm’s global academic


partnerships. Impact Forecasting combines the latest findings of Aon Benfield Research with our own modelling technology to provide cli- ent-specific


solutions. For example, on our


forthcoming European Wind storm model, we are working with the Uni versity of Cologne whose windstorm research will be directly used inside the model. This includes the statistical and probabilistical downscaling of windstorm events data to a much greater resolution, from approximately 100km x 100km to 1km x 1km. We frequently sched ule face-to-face discussions and joint working sessions with our academic partners on key risk topics, which helps to develop long-term relationships and deliver the most rel evant research to the insurance industry.


www.intelligentinsurer.com


25.10.11 TUESDAY


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20