Next-generation capital models

11 October 2017

Interest in economic capital modeling has been growing, driven by regulators and ratings agencies, as insurers look to make ever-more efficient use of their capital. By Stuart Collins

Insurers are operating in an uncertain world. Large and often unexpected losses, as well as increased volatility in investment markets, have seen insurers’ interest in enterprise risk management (ERM) and modeling grow.

Many now recognize the need to go beyond underwriting and investment, and take a much broader view of risk and the volatility of earnings.

Stochastic modeling

Stochastic capital modeling tools, also known as economic capital models (ECM) or dynamic financial analysis (DFA), have become an important tool for many insurers.

According to Micah Woolstenhulme, Head of Risk and Economic Advisory at JLT Re, ECM takes financial planning to the next level.

“With economic modeling, insurers can quantify the likelihood of certain outcomes for a wide range of risks, helping management to understand the potential volatility for earnings or premium income in a given scenario,” he says.

Regulatory driver

Stochastic capital modeling tools have been around in the insurance industry for more than a decade. But interest has been growing of late, driven by regulators and ratings agencies, as well as an evolution in insurers’ risk management and changes in the risk environment.

The National Association of Insurance Commissioners, the body that drafts regulatory standards for state insurance supervisors in the US, has placed an increasing focus on ERM in solvency regulation.

In 2015, US regulators introduced the Own Risk and Solvency Assessment (ORSA), which requires US carriers to assess their current and future risk through an internal risk self-assessment process.

In Europe, Solvency II not only requires insurers to adopt an ERM approach, but it also potentially rewards them for doing so.

The EU capital and risk management rules allow insurers to use their own internal models (a form of economic capital modeling) to calculate regulatory risk, which is often favorable when compared with the standard model set out by regulatory authorities.


Adding Value with economic capital models

  • Profitable growth
  • Pricing and portfolio management
  • Enterprise Risk Management
  • Franchise value analysis and optimization
  • Advising rating agency process
  • Supporting solvency regulation



Next phase for stochastic modeling

Rating agencies have placed increased emphasis on an insurer’s ERM in recent years, and bold moves are afoot that are likely to further raise the profile of stochastic modeling in coming years.

A.M. Best, for example, is modernizing its Capital Adequacy Ratio (BCAR) process, incorporating stochastic modeling into its ratings process.

The ratings agency plans to use stochastic modeling to inform its views on capital adequacy, modeling a range of potential risk outcomes and confidence levels.

Insurers’ competitive advantage

In addition to supporting solvency regulation and credit ratings, ECM also creates competitive advantages for an insurer, with potential applications in informing strategy, risk profiling, and risk mitigation.

For example, franchise value analysis provides analysis of earnings and can assist decisions on efficient reinsurance structures.

ECM can also feed into strategic analysis in areas such as profitability, market entry, growth opportunities, competitor analysis, and asset and credit risk.

Accessibility for capital models

Increases in computing power and advances in analytics have made economic capital modeling more accessible and relevant to insurers.

Insurers are increasingly using sophisticated models, and not just for catastrophe risk but also for actuarial, claims, and investments.

Each of these components has undergone its own evolution, and combined they are helping make economic capital modeling the powerful tool we see today, according to Woolstenhulme.

“Economic capital modeling is gaining steam as each component has become more sophisticated and requires a capital model to bring them all together under one framework,” he says.

Next generation of capital modeling platform

JLT Re is developing the next generation of its capital modeling offering.

“This new software will be unique to the market and exclusive to JLT Re clients, combining leading technology with JLT Re’s own approach to loss simulation models and experience with catastrophe modeling and advisory,” says Woolstenhulme.


Strengths and weaknesses of economic models

Economic models are particularly good at handling the main risks faced by insurers, many of which are already well understood.

These are namely insured risks (such as natural catastrophes), economic and financial market risks (for example, interest rates, bond yields or equity prices), as well as contractual risks (such as reinsurance contracts), as well as the performance of non-catastrophe losses and claims processes.

“Economic capital modeling is the beautiful marriage between structured and stochastic risk modeling. Structural models, like catastrophe models, are based on scientific expertise and past events, but stochastic modeling helps fill in the gaps with actuarial experience,” says Woolstenhulme.

Economic modeling may not be suitable for certain operational risks, where risk may be plausible but impossible to assign probability with reliability. However, such operational risks can be dealt with by stress testing, which will at least give some insight into the potential magnitude of loss.


Interpreting models

Yet, developing an ECM is not just a “plug and play” exercise, according to Woolstenhulme.

Designing and implementing a capital model is a bespoke exercise that requires the insurer to know what questions they would like to see answered, he says.

“The expertise of the practitioner is everything when it comes to bringing capital models to bear on key decisions,” explains Woolstenhulme.

“It is critical to have a partner with expertise in building and validating capital models, and that is able to interpret model output and understand the assumptions and methodologies underpinning the model,” he says.

Please contact Micah Woolstenhulme on +1 215 309 4637 or micah.woolstenhulme@jltre.com

Download button