A Stabilising Effect: The ICAS Framework In Action

It is two years since the Financial Services Authority ("FSA") has required insurance companies to calculate capital requirements under the Individual Capital Adequacy Standards ("ICAS") framework. The nonprescriptive nature of the regime means the industry is still getting to grips with producing capital assessments and best practice is continuing to emerge. Deloitte conducts a comprehensive ICAS survey of the life insurance industry on a regular basis. This survey gives valuable insights into how companies are approaching the new regulatory requirements and what they see as the key remaining challenges. With the European introduction of Solvency II planned in the next few years this is only the beginning of a journey for the UK Life Insurance Industry.

Background on ICAS

Since the beginning of 2005 the FSA has required insurance companies to calculate capital requirements under the ICAS regime.

The capital held by a firm must be sufficient to withstand a range of possible adverse events within the broad categories of market, insurance, credit, liquidity, operational and group risk. The strength of these events should be at the level that might only be seen once in every two hundred years, translating to a 99.5% confidence interval over one year.

The FSA has taken a principles based approach to the introduction of ICAS. Over the last year the FSA has communicated on ICAS matters in various ways, including several newsletters, a sector briefing and presentations at conferences.

However, it remains difficult for firms to anticipate the FSA’s expectations from an Individual Capital Assessment ("ICA") submission and understand how an appropriate balance will be struck between flexibility and consistency when the FSA reviews firms’ work. With limited resources firms must decide which areas of the assessment to focus on in order to achieve a robust and effective ICA which is acceptable to the FSA.

The FSA has an ongoing programme for reviewing firms’ ICAS assessments and an increasingly large number of life insurance companies have received Individual Capital Guidance ("ICG"), reflecting the FSA’s view of the capital required to cover the risks faced by the firm. In nearly all cases the FSA has recommended ICG in excess of the firm’s view, although not in excess of actual capital available.

The majority of firms continue to apply stress tests to an opening balance sheet. We are surprised that more firms have not sought to develop models enabling them to carry out fully realistic projections over periods of one year or longer especially as survey respondents continue to report the desire to do so. It is only when companies are able to investigate capital requirements over longer periods that the principles of the ICAS regime can be more usefully extended to business planning.

The correlation matrix continues to be the primary source of calculating diversification benefits and the majority of companies are now supporting this with various forms of scenario testing and consideration of tail correlations and non-linearity.

For market and credit risks there is increasing conformity in the strength of the stress tests applied. Where there are significant differences these tend to reflect the nature of the actual assets held or include adjustments to take into account the individual company’s methodology. There continues to be wide variations in the insurance risk stress tests but differing approaches to aggregation of the risks can make comparisons difficult. We believe that most companies are now comfortable with their individual stress tests and will in future focus resources on developing more sophisticated methodologies and scenarios.

Operational risk

With the introduction of Solvency II not far away, it is perhaps worrying that operational risk continues to be one of the more challenging elements of the ICA. This is reflected in the variety of approaches used by the survey respondents. We anticipate that as more industry and firm-specific data becomes available over time, operational risk assessments will become increasingly sophisticated, with improved capture of the underlying risk data and allowance for the mitigation of capital requirements through proven processes and controls.

Most companies have documented operational risk policies and many now have processes in place to capture and categorise internal operational losses. We anticipate that the others will use the ICAS framework to introduce formal policies and mechanisms to capture losses. Fewer respondents are able to capture near misses. This is important in order to identify actions to prevent the event recurring and incurring costs.

The FSA recognised in its recent ICAS Sector Briefing that it is difficult to quantify operational risks at the required confidence interval. However, they are firmly of the view that firms should be able to articulate how their methodologies have been derived and how judgments have been formed.

Firms should also be able to demonstrate a clear link between the operational risk assessment in the ICA calculation and their day-to-day risk management. This means that a bottom-up approach to identifying and quantifying the risks is more appropriate than a simplified top-down approach, where, for example, operational risk capital may simply be a percentage of capital requirements from other risk categories.

It is unlikely that all adverse events will occur simultaneously at the 99.5% level and it is reasonable that there will be some level of diversification benefit between the items of risk represented by the individual stress tests. Diversification benefit is typically the largest item within an ICA assessment. The majority of survey respondents (87%) indicated that they use a correlation matrix as the primary method for aggregating capital requirements and calculating diversification benefits. Around two-thirds of the insurers we surveyed also use scenario testing as a tool for calculating diversified capital requirements.

The correlation matrix approach has the advantage of being transparent although it can be very difficult to derive the individual correlation assumptions and consequently have confidence that the resulting diversification benefit is reasonable. Additionally, the correlation matrix will not capture any non-linearity between risks (where the capital requirements from risks occurring together are proportionately larger than the sum of the individual items) and could lead to double-counting of the impact of management actions.

Scenario testing can offer a solution to these problems as all risks within the scenario are assumed to occur simultaneously. In particular, carefully designed scenarios will automatically capture any non-linearity between the underlying risks. Although scenario testing brings its own challenges, namely how to identify a combined scenario which represents a 1-in-200-year event, a well presented methodology, and comparisons to the correlation matrix approach, demonstrates sound thinking to the FSA.

Management actions often constitute a significant contribution to the ICA result, especially for with-profits business. In some circumstances, differing views on the strength of a 1-in-200-year event, or the amount of diversification benefit, may be offset by the extent to which management could take mitigating actions. It is important for firms to articulate the management actions that they could take under extreme scenarios and to explain in the submission to the FSA how the modelled actions are in line with those agreed by the board and stated in the Principles and Practices of Financial Management ("PPFM"), which sets out how any with profit funds will be managed. Firms should also be able to demonstrate how the actions remain compliant with Treating Customers Fairly ("TCF") principles in those situations. The impact of management actions can be investigated by running the scenario twice; once with the actions in place and again where they are ignored.

The holy grail of scenario testing is to identify those scenarios to which the business is particularly exposed, whether this is for ICA purposes or to satisfy the firm’s internal risk appetite. Although these scenarios target the same overall confidence interval, the strength of each constituent may represent different percentiles on their individual distributions.

The following chart shows the average split of capital requirements between the six broad risk categories before diversification benefit.

This chart is very similar to that shown in the previous ICA Survey, suggesting that a reasonably stable position has been reached across the industry. The split for some companies will change over time as the ICA leads to additional actions being taken to manage the risk profile.

A comparison of ICA requirements with those under Pillar 1 suggests a 50% increase in the capital needed to support the business, on average, although this is offset to some extent by a higher value of assets under Pillar 2. It should be noted that this does not include ICG and, with the first ten reviews generating ICG ranging from 110 to 170% of the ICA, this should not be underestimated.

In due course we expect capital ICG as a percentage of ICA to reduce as fewer companies overlook items of risk, methodologies are developed, best practice continues to emerge and feedback from the FSA is taken on board. Recent anecdotal evidence suggests that this may already be the case.

In our previous ICA survey we saw the debate on ICAS move from discussions on the proposal to how the calculations themselves were to be performed. In the 2006 ICA survey we have seen the debate shift to how to embed capital calculations and risk management processes into the business; what the FSA calls the "Use Test". Experience from Basel II in the banking industry demonstrates that the challenge this presents should not be underestimated and Solvency II will further drive the need to firmly embed risk and capital management practices.

Where firms have ensured that ICA is more than simply an actuarial exercise, they are now beginning to reap the rewards of their hard work and have started embedding risk identification, monitoring and management processes into elements of their day to day business. For example we have seen a significant increase in the extent to which ICA is influencing product pricing although is it clear that most companies have yet to fully embrace using ICA in this area (around one third of companies now compare product profitability with capital requirements calculated under the ICA). It was also clear from the study that just 10% of companies’ business plans are yet to be influenced by the ICA. We believe that as familiarity with the impact of actions and decisions on capital requirements grows, the ICA will play an increasingly important role in formulation of strategy and in making key business decisions.

In the relatively short period since our previous survey, we have seen some significant changes in the extent to which ICA is influencing the management of life insurance companies. We also anticipate that for many companies the ICA results will contribute to hedging strategies, product pricing developments and the future new business mix (to balance diversification benefits and efficient use of capital). For some of the larger insurers the ICA may also lead to considerations of appropriate fund structures (eg by merging funds). Although becoming business as usual, firms clearly see significant benefits in further developing their ICA methodology and embedding the principles of risk management within their firm. Incentives include a potentially lower yet more robust ICA, further insights into the risk and cost drivers of the business, more efficient use of capital and being better prepared for the requirements of Solvency II, which are fast approaching.

The latest in our series of surveys tracking how companies are responding to the new capital assessment requirements was completed by Deloitte in mid 2006. 38 UK life insurance companies participated in the survey covering 74 separate funds representing approximately 80% of the UK life insurance industry.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.