April 4, 2017
Changing Business Models, ‘New’ ERM
by Henry Essert
Here are three key developments that insurers should incorporate into their evolution on enterprise risk management (ERM).
Significant social, technological, economic, environmental and political forces are reshaping the needs and expectations of insurance buyers, as well as the business environment in which insurance providers operate.
Even a partial list of these forces is daunting: aging populations in developed markets; different needs and purchasing behavior of younger buyers of insurance; self-driving vehicles; telematics; artificial intelligence; the internet of things; and persistent low interest rates. With so many forces in play, it’s difficult to determine the exact landscape of the new insurance world.
But it’s not too early for insurers to prepare.
Regardless of exactly how they plan to address a rapidly changing and more unstable world, one capability that will remain critical to all insurers’ success is enterprise risk management. We describe below three key developments that insurers should incorporate into their ERM evolution.
Insurers’ business models are changing and ERM needs to keep pace.
Stress testing will join economic capital as the main risk decision tool.
VAR-based economic capital measures originated in banking and asset portfolio management more than 40 years ago. Over the last couple of decades, the insurance industry has widely adopted the concept.
This is particularly true for insurers’ credit and market risk taking, areas where the VAR concept is endemic. For some aspects of insurance risk, like statistical variability around a stable mean, the concept also fits well. In an insurance world where credit, market and insurance are insurers’ main risks, economic capital is effective. But what if the world changes to one where other risks join these at center stage?
Life insurance in a persistent low-interest-rate environment with rapidly evolving distribution models provides a clear example of recent change and its implications for ERM. The bulk of many life insurers’ liabilities and supporting assets are composed of permanent type products they wrote when asset returns were markedly higher. These higher returns supported the stable distribution model of a sales force based on up-front commissions. In turn, this fit the products’ complex features that needed such a model to explain and sell them. Delivering on these guarantees necessitated focus on the credit and market risks they created. And VAR was developed to manage these risks.
However, now that asset returns are much lower, supporting this distribution model will be difficult. Fortunately, other less costly models are available and probably preferable to younger buyers of insurance. This demographic group has shown a preference for a more t-to-purpose protection model that is less permanent and less complex. As a result, credit and market risks cease to be ERM’s overwhelming focus. Instead, strategic and operational challenges created by transitioning to and maintaining the new business model take center stage, as do the risk tools that can address these challenges. Among these, stress testing figures most prominently.
Trends in the property and casualty sector also point to a shift in risk focus and risk management tools. Impending and actual changes in the nature of driving and vehicle ownership will radically and permanently alter the auto insurance landscape. Developing an understanding of the implications of these changes and their risks to an insurance enterprise needs a tool like stress testing. Similarly, an increased emphasis on assisting customers with mitigating and managing their own risks, rather than just insuring them, moves more of an insurer’s risk profile out of the traditional risk-taker role and into a service provider model. VAR is a good risk tool for a risk taker, but stress testing is the tool best suited to the service provider model.
Lastly, we note that rapidly emerging technologies, often cited for their role in shaping customer preferences, also shape insurers’ own capabilities. Insurers have begun to modernize their back offices, and computing power continues its exponential growth. Operational challenges and resource demands to implement new and improved risk tools, like stress testing, will diminish significantly. With benefits going up and costs going down, it seems clear that stress testing is on its way to a prominent ERM role.
Customer analytics decision platforms will become the key focus of model risk management efforts
Model risk management (MRM) is receiving extensive ERM focus at present. Much of the original impetus may have come from European companies seeking to validate their Solvency II internal models. In the U.S. and Canada, due in part to direct or indirect regulatory encouragement, the scope goes beyond economic capital and solvency models, and most insurers seek to apply their efforts to all models.
The early priority for validation has skewed toward economic capital and complex liability valuation models. Insurers with advanced MRM capabilities have begun to focus more attention outside of risk and financial reporting models. This is to be expected to some degree, as insurers model validation activities work their way through their inventory of models. In addition, as they develop a working experience of risk rating their models, many are reconsidering the irrecoverable nature of product pricing decisions and the importance of getting those models right. In other words, while small errors in financial and risk reporting models can be rectified once errors are uncovered, losses from inadequate premium charges are permanent.
The impetus for higher attention to pricing and risk selection models is further amplified when insurers implement newer, non-traditional approaches. Without a long history of successful use, newer customer analytic models put a higher priority on their timely and thorough validation. Additionally, we have observed insurers further enhancing their level of attention when these models move to autonomous execution mode. In this mode, the model makes decisions in an automated fashion without manual intervention or deliberation. Deploying more models of this sort is a common feature of most visions of the near-term future of insurance. As their use expands, so too should ERM’s focus on effective risk management of these models. In an environment in which these types of customer analytics decision platforms become an insurer’s key business engine they also will need to become the key focus of MRM efforts.
Small errors in financial and risk reporting models can be remedied; however, losses from inadequate premium charges are permanent.
Risk diversification measurement will become the single most important element in economic capital calculations
There is a continuing focus on the effectiveness of economic capital modeling, especially in connection with IAIS and regulatory efforts outside of the U.S. In the U.S. as well, insurers continue to look at how they can improve their calculations. However, one area we believe attracts insufficient attention is diversification.
Not only is an effective understanding and quantification of diversification an important goal in the current insurance environment, it will likely become even more critical in the future. As the new risk profile moves away from a credit/market nexus to a more diverse insurance, business and strategic risk set, managing the interaction between and among them will be especially important. If customers move to a more holistic view of insurance and blur the distinctions between life, property and casualty and health, just quantifying the diversification across all insurance risks will be a key task on its own.
If they haven’t done so already, CROs should start to sketch out a few versions of what their company might look like in the future and consider what might be required of their ERM capabilities. They can adjust and clarify this high-level road map as the future becomes clearer.
Considerations CROs should keep in mind while creating this roadmap include:
- On the life side in particular, credit and market risks will cease to be ERM’s overwhelming focus, but stress testing will figure more prominently in new business models.
- Assisting customers with mitigating and managing their risks instead of just insuring them will move more of an insurer’s risk profile out of the traditional risk-taker role and into a service provider model. VAR is a good risk tool for a risk taker, but stress testing — which is becoming cheaper and easier to do — is better suited to the service provider model.
- As advanced customer analytics decision platforms become an insurer’s key business engine, they will need to become the key focus of model risk management efforts.
- As insurance becomes more holistic for customers, quantifying diversification across all insurance risks will be a key task for insurers.