In today’s economic climate and with the increasing convergence of insurance regulation on a global basis, insurers face growing risk challenges. These include optimising economic and regulatory capital, managing risk effectively across the enterprise and providing faster, higher quality financial reporting through improved data integration.
The positives of the internal modelling process were stressed in the recent Global Risk Management Webcast, presented by Reactions and sponsored by IBM.
To view the replay of the webcast for free, please visit here.
To download the presentation from the webcast, clickhere.
Curt Burmeister, vice-president of risk solutions at Algorithmics, an IBM company, said that as well as producing immediate results, the process of producing an internal model can generate many other benefits.
Burmeister said that as well as being an effective way of visualising all of an insurer’s risks on one balance sheet, a single internal model can also help to make business decisions and can be used as a presentation tool for shareholders and rating agencies, in addition to the benefits under Solvency II.
“It’s a consistent view of capital across the organisation,” he said. “It’s one model where everybody has their input.”
Internal models can also lead to reduced capital requirements, Burmeister said, typically because firms are “able to relax some of the conservative estimates that you put in when you aggregate apital,” and also generate competitive advantage for the firms concerned. However, he stressed that the building process itself may have one of the greatest benefits in stimulating discussion about what a given insurer’s risks are and how they can be modelled.
“That can be quite contentious at times, but it starts to promote a risk culture across the organisation,” Burmeister said.
He conceded, however, that such a process can prove costly, both in terms of time and the people who must be involved, making it an unattractive prospect for insurers.
“These are not simple problems to solve, for sure, and there is often quite a bit of debate about what the best approach is,” he said. “It’s a very significant investment from a people perspective.”
Indeed, there can be significant challenges to such a process, Burmeister says, including simply organising the various teams that will contribute to the project so that it is clear where various roles and responsibilities lie.
“There can be differences of opinions about the best way to go forward and overcoming that organisational challenge is the first hurdle that they go through,” he said.
Equally, Burmeister says that the need for strong, reliable data means that substantial commitment is needed from IT, risk management and actuarial departments.
“Of course with the challenges come opportunities...and not everybody will see all of these, but you do end up with a much better data story once you’ve gone through those exercises because the demands of the model require it,” he said.
IBM has also noted that operational risk is proving a big challenge for many insurers as part of their modelling process, not least because it marks a substantial difference to more quantitative types of modelling.
Gordon Burnes, head of worldwide marketing for IBM’s risk analytics business, said that many insurers may have built up more of a silo approach to the many risks that are defined as operational.
“That makes it hard to get the full range of insight needed to really manage the full range of operational risks,” Burnes said. “The fragmented approach, not only is it very hard to aggregate information from across the company, but it’s very expensive.”
As a result, he says that many have begun working on more integrated approaches to operational risk that allow for the required inputs to an internal model, while also standardising risk management across a firm.
“There’s real value in taking a holistic approach to deploying a operational risk management system as it does operate in isolation to other risk management disciplines,” said Burnes. “For instance, without streamlining overall efforts, business process owners get overwhelmed with requests for assessment information and data from internal audit, risk management, compliance and even IT.”
Burnes suggests that without such efforts, individual data requests can each be handled slightly differently, resulting in a drop in overall data quality, as well as response times for collecting the data. What’s more, he says that such an option can prove to be very expensive, largely because companies have to support multiple systems and data collection processes around the organisation.
Nonetheless, the investment can be fruitful. Burnes likens the state of operational risk management to that of customer relationship management (CRM) in the past.
“Before CRM systems were deployed to provide a single view of the customer, customer information was scattered throughout the enterprise, with marketing, sales and support all having different views of the customer,” he says.
“In many companies today, the same is true with risk. The risk management, compliance, audit and IT groups all have different views of risk.”