Benefits of Scenario Analyses

A scenario can be deterministic or stochastic, historical or hypothetical. An analysis based on deterministic scenarios typically considers just a few scenarios. These scenarios might be historical or hypothetical; each has its advantages. Historical scenarios are easier to design, but hypothetical scenarios are more flexible.

Stochastic scenarios are based on simulations — typically between 1,000 and a million — and are driven by adjustments to the parameters and input variables, based on their historical distributions and co-variances.

Scenarios can be driven by events or problems that may arise from a company's portfolio of business lines and risks. Event-driven scenarios are based on plausible events and how they might affect a firm. Portfolio-driven scenarios are driven by the vulnerabilities of the portfolio held by a firm.

Though scenarios tend to be mostly quantitative, they can be used to develop alternative views of the future that are meant to stimulate thinking about how a company might best respond to fundamental changes in the business environment. These scenarios, developed through an internal and external review process, are qualitative and descriptive.

Limitations of Scenarios

Scenario analysis is subject to weakness in three major areas: the quality of the model, the quality of the data, and the quality of the scenario team.

Models may be mathematical and structured, or implicit and intuitive. Mathematical models are fairly straightforward in that they are clearly and explicitly specified. For mathematical models, the quality of the model may be affected by specification and parameter errors.

Data quality also affects the reliability of the model. If data is of poor quality, the result will likely be incorrect or unreliable. Lack of data may also cause models to be unreliable. For example, historical loss data for virtually all emerging risks, such as nanotechnology, is lacking, which makes it nearly impossible to use quantitative models.

Third, the quality of the scenario analysis team is essential. Scenario analysis is a valuable part of a risk manager's toolkit. For scenario analysis to be effective, a certain level of expertise and solid business judgment are needed. The staff who use the models and subsequently analyze, review, and interpret the results must be highly qualified. Familiarity with the model and data — and their limitations — is important.

If an implicit model is used, the quality of the staff is probably even more important. Care must be taken to clarify all steps for later review and analysis, and for varying the assumptions to generate further scenarios.

Finally, scenarios should be used efficiently. Like other business techniques and processes, scenarios should not be overused. The cost of developing, maintaining, and running scenarios should be weighed against the benefits.

Building a Scenario

Before attempting to build a scenario, it is important to identify the set of scenarios that are relevant. Only the most probable scenarios with the highest impact on the company/business line/risk should be evaluated. There are several steps involved in building a scenario that facilitates business decision-making, including:

1. Define the risk or issue.

2. Recruit a team with appropriate skills to analyze the scenario.

3. Identify the factors affecting the risk.

4. Obtain and review the data/information needed.

5. Develop a methodology for how to technically answer the questions.

6. Document the results of the analysis, making explicit all assumptions and how they were derived.

7. Set up a peer-review process for verifying the robustness of the methodology, models, and procedures.

8. Use the results of the scenario analysis to make a business decision.

As discussed, there are many practical applications for using scenarios, including risk management; strategy and planning; underwriting and pricing decisions; and capital management. While these are typical uses of scenarios, they represent only some of the ways scenarios can be used.

Scenarios are particularly important for strategic planning decisions. Through scenario analysis, management is able to evaluate the financial implications of alternative responses to potential external market events before they occur. This enables them to be better prepared and to act quickly and decisively in the event the scenario occurs.

Insurers also use scenario analysis for underwriting a wide range of risks. Underwriting decisions are made after building a small model and running Monte Carlo simulations by varying the parameters of the model. Of course, models are not perfect at specifying future outcomes, so the skill, judgment, and experience of the practitioner remains very important.

Scenarios can also be used to improve capital management efficiently, which ultimately improves the return on equity. In addition, they can be used to determine capital adequacy. To attain its desired level of financial security, a reinsurer requires a certain amount of capital. Scenarios can be used to estimate the minimum level of capital required for the designated level of security.

Scenarios can also be used to support capital-management decisions. An optimal reinsurance program as well as hedging needs can be derived from scenario analysis. Furthermore, equity issues and stock buybacks, the creation of sidecars, securitization of risks, and dividend payment decisions can be facilitated by using scenarios.

Developing State-of-the-Art Analysis

Though the use and sophistication of scenario analysis in insurance has improved, it is far from perfect. The industry is not yet fully applying state-of-the-art standards. State-of-the-art scenario analysis includes models and processes that take risk-adjusted returns, assets, liabilities, capital, and other key features into account. The state-of-the-art for insurers includes excellence in the following types of scenario analysis. This is adapted from Standard & Poor's (2006) Insurance Criteria: Refining the Focus of Insurer Enterprise Risk Management Criteria:

? A scenario-based process for assessing and testing the risk-adjusted returns for all major business lines.

? A global model of assets and liabilities that can be stress tested with insurance, economic, and financial market shocks.

? A scenario-based process for assessing the optimal use of capital, including stock buy-backs and dividends.

? A regular program of internal scenario tests related to shocks, such as natural catastrophes and pandemics, and also shocks on each major asset class and business line.

? A regular program of internal scenarios based on economic and financial market shocks.

? A department tracking the emerging risks and a qualitative approach of creating scenarios related to these risks.

? Qualitative scenarios for operational risk management or other qualitative risks, including workshops featuring brainstorming on potential operational risk scenarios.

Benefits

In addition to the benefits mentioned earlier, insurers can use scenarios for enterprise risk management, which in turn can improve the overall financial performance of the firm. The process of conducting a scenario analysis can even be useful for learning more about the effectiveness of an insurer's risk management practices and the environment in which it operates. Moreover, scenarios can be used to communicate with key stakeholders, For example, scenario analysis helps insurers comply with regulatory requirements, increases the likelihood of receiving a higher rating from rating firms, and clarifies the insurer's risk to investors.

For a number of reasons, including the current financial crisis, the use of scenarios is likely to increase along with improved technology, competitive pressures, and the increased oversight capabilities of regulators and rating firms.