Unlocking the Value in Economic Capital
Unlocking the Value in Economic Capital
by Sim Segal and Mike McLaughlin
How can insurers expand the reach of their efforts with regard to Economic Capital? Some experts contend that the answer can be found in shifting the focus towards value.
Enterprise Risk Management (ERM) is a popular topic these days. Many insurers are in the process of establishing some form of ERM program or are well along the implementation path and are refining their approach. A key component of ERM is the risk assessment phase, which, for most insurers, takes the form of Economic Capital (EC). EC is the process of determining the amount of capital needed to survive based on sophisticated risk modeling. EC has the potential to shed new light on valuable opportunities for insurers, and some have reaped these benefits.
However, many insurance companies have not fully embedded their EC-based risk assessment into the broader ERM process. A good diagnostic test for this is to identify the extent to which decisions throughout the company are supported by EC metrics. For example, at some insurers, the only decisions supported by EC metrics are certain decisions at the corporate level, e.g., capitalization levels, reinsurance retention levels, etc.
So, how can insurers facing this issue expand the reach of their EC efforts? Based on Deloitte Touche Tohmatsu (DTT) global experience in EC and ERM engagements, and on analysis of industry practices, DTT believes that to fully leverage their EC programs, insurers must shift the focus towards value. This discussion document suggests ways in which this can be done.
Shift the Focus Towards Value
Most EC programs involve a fairly robust effort to:
- Identify key risks, including insurance, market and credit risks.
- Define the shape of each risk curve (outcomes/probabilities) and risk correlations.
- Develop stochastic scenarios.
- Project cash flows and capital flows over the time horizon.
- Translate the projection into income statement and balance sheet impacts.
- Define survival (e.g., GAAP capital at or above a given level) and confidence level (e.g., survive 99.5 percent of scenarios).
- Determine the Economic Capital–the internal required capital–as the capital needed to survive with the defined level of confidence.
While this effort produces a large amount of information, the end result–determining EC–focuses on the extreme tail end of the results set (see Chart I, Area A). However, after the EC is calculated, the remainder of the curve, particularly the large data set near the mean, remains unused (see Chart I, Area B).
This is unfortunate, since this part of the curve contains valuable information. This is the area in which there are more frequent occurrences of risk events, represented by volatility near and around the mean for each dimension of risk–mortality, yield curves, etc. Therefore, it is in this area that there are more opportunities to take action, i.e., to manage the risks. In addition, this portion of the data is more credible than the data upon which EC is based. The middle portion of the curve is based on a larger set of data. In contrast, the further out one goes on the tail end of the curve, the fewer the data points there are that support the shape of the curve.So, how can this information be used? By shifting the focus from capital to value. In the process of constructing the risk scenarios and resultant financial projections to arrive at EC, most of the necessary information has already been produced to calculate value, in the form of discounted distributable earnings (see section on Calculating Value). Certain enhancements are needed to bridge the gap between EC and value and these are discussed here.
Once these enhancements are completed, a value–based approach can be supported. The expected (baseline) value of the company is produced, as is a portrait of how that value is expected to vary, based on the existing risk profile (see Chart II on page 32). This switches on a direct current between EC and decision–making processes, since value is the language of decision–making. All major decisions–strategic planning, pricing, etc.–can be evaluated by examining their potential marginal impact on this enterprise risk profile expressed through "value–volatility." Each decision is supported with the information of how much it is expected to increase value and also, perhaps more importantly, ranges of scenarios (includes their likelihood) around that single point estimate. In addition, the type of information provided in Chart II can be used to inform one of the most important tasks in ERM - managing the enterprise risk exposure to within the risk appetite, as expressed by enterprise value volatility.
Enhancements: Building the Bridge Between EC and Value
There are three key enhancements needed to convert the EC model to a value–based model:
- Project new business.
- Include operational and strategic risks.
- Project required capital.
Project New Business
Some EC models include future new business projections. However, many EC models do not. To calculate value, the model must have a projection of future new business consistent with the strategic plan, over the chosen time horizon. To the extent that the strategic plan's future new business projection does not extend to the end of the chosen time horizon (and more than likely it does not), a reasonable assumption must be developed to project the future new business beyond the strategic plan period.
Developing a baseline future new business assumption, particularly beyond the strategic plan period, carries with it a high level of uncertainty and can cause those developing the assumption some discomfort. It often helps to realize that more important than the absolute level of the baseline assumption is the shape of the volatility curve around that baseline. In addition, some additional comfort can be gained by doing a reasonability check–calibrating the value calculation to market value to determine the market's implicit assumption as to the expected level of future new business.
Include Operational and Strategic Risks
Many insurers use EC models that include stochastic scenarios that reflect market risk, credit risk and insurance risks. However, the scenarios behind most EC models usually exclude operational and strategic risks. Operational and strategic risks include such risks as poor strategic planning or strategy execution, litigation, fraud, reputation risk, etc. These risks are significant, particularly in terms of how quickly they can destroy a company's value. Therefore, operational and strategic risks must be incorporated into the model to quantify the impact on value.
One reason why operational and strategic risks may have been ignored up until now is the fact that EC models do not always adequately reflect future new business. Without this, it can be difficult to adequately capture certain of these risks. For example, quantifying the impact of poor strategy execution (e.g., the risk that only half the planned growth will be achieved) can only be quantified if baseline growth and volatility around that baseline are included in the model.
Another reason insurers limit the incorporation of operational and strategic risks may be that few are aware of effective methods to develop operational and strategic risk scenarios. One such technique is to develop deterministic scenarios using a Failure Modes and Effects Analysis (FMEA). FMEA is a method adapted from manufacturing industries, where an internal analysis is conducted to determine where a process could fail, how it could fail and what could be the impact of the failure. There are challenges in applying this approach. It takes experience in the FMEA process and a disciplined approach to get meaningful, consistent scenarios across the enterprise. However, through an interactive exercise between internal subject matter specialists and corporate ERM, this technique can produce he required credible scenarios in a short time frame.
This approach to incorporating operational and strategic risks into EC modeling has advantages over alternatives currently being used by many insurers. Unlike the stochastic approach, it does not heavily rely on large amounts of external data, which may either be unavailable or inapplicable to the company's true operational and strategic risks. Also, unlike a fixed percentage add–on (e.g., 15 percent), this approach yields a dynamic capital amount that increases or decreases with the level of risk, e.g., if mitigation activities are enhanced, then the EC amount is reduced.
Project Required Capital
EC models project actual capital amounts over the time horizon. However, future changes in required capital amounts are needed for the value calculation. The most accurate method of determining the future required capital amounts is to perform another EC calculation at each future time period for each risk scenario–this is what stochastic–on–stochastic projections are designed to do. Unfortunately, these calculations can often become prohibitive in terms of run time. Therefore, it may be preferable to use a shortcut method. One such shortcut method is to develop a factor–based formula that reproduces EC at time zero and that can then be used as a proxy for EC in the projection. In the United States, this is analogous to determining the (pre–C3 Phase II) NAIC RBC level at time zero that would support GAAP capital equal to EC.
To get the most out of their EC and ERM programs, insurance companies can leverage their EC models to support a value–based approach, making the connection between risk and value. This can help infuse a company's decision–making processes with the risk intelligence needed to enhance enterprise value.
1 Two other items are needed to perform the value calculation, but were not discussed in depth here. 1) Statutory financial projections: This was considered a minor item. Statutory financials are often already available as a basic output of EC models. When this is not the case, it is a fairly trivial matter to produce the required Statutory adjustments. 2) Risk discount rates: While the value calculation shows the weighted average cost of capital as the discount rate, it may be appropriate to use different risk discount rates for blocks of business with varying levels of risk. However, appropriate risk discount rates are often available from other applications in the company, e.g., hurdle rates developed for risk–adjusted return measures such as Return–on–Equity.
Mike McLaughlin is leader, Global Actuarial and Insurance Solutions. Deloitte Consulting LLP. He can be contacted at email@example.com
Sim Segal is senior manager, Deloitte Consulting LLP. He can be contacted at firstname.lastname@example.org
Reprinted with permission from Deloitte Consulting LLP.