Why Insurers Care About Economic Capital
Economic capital affects many areas of insurance company operations, including setting risk appetite and retentions, allocating capital across statutory entities and business segments, measuring performance, determining investment philosophy and allocation, adjusting product pricing, and mix of business. Capital inadequacy can lead to rating downgrades by rating agencies and possible regulatory actions.
The National Association of Insurance Commissioners (NAIC) passed a model act1 that applies to any individual U.S. insurer that writes more than $500 million of annual direct written and assumed premium, and/or insurance groups that collectively write more than $1 billion of annual direct written and assumed premium. This is subject to adoption by each state.
Modeling economic capital provides the foundation for the determination of risk capital. Smaller insurers are encouraged to calculate economic capital as a matter of best practice and for enhanced ratings by A. M. Best Company and Standard & Poor’s Insurance Rating.
Do Mid-Sized and Small Insurers Need to Perform Economic Capital Modeling?
Economic capital modeling is comparable to catastrophe modeling for property insurers: you may not be required to do it, but it is the smart thing to do. Many insurers perform catastrophe modeling to assess the risk in their portfolio of exposures, to guide future underwriting strategy, and to determine their optimal reinsurance structure. Those benefits are too important to ignore, so catastrophe modeling has become an industry standard that is typically performed annually by an insurer or an insurer’s reinsurance broker. Similarly, economic capital modeling for mid-size and small insurers is not required at this time, but is the smart thing to do. Economic capital considerations should guide many underwriting and investment decisions.
How Economic Capital Modeling Works
Economic capital modeling uses a probability-based scenario generator to estimate the future financial results of an insurance organization. Detailed information about an insurer’s operations is entered into a modeling package and the package produces alternative versions of the insurer’s financial statements for prospective years. Key risks modeled in an ECM include underwriting risk, reserve risk, liquidity risk, natural catastrophe risk, asset risk, reinsurance credit risk, and operational risk.
Economic Capital’s Effect on Insurers’ Company Ratings
For all major rating agencies, capital adequacy is the most important factor in your company’s financial stability ratings. A.M. Best Company, Standard & Poor’s Insurance Ratings, Moody’s Investor Service, Fitch Ratings, Weiss Ratings and Demotech all use different methodologies to determine rating levels, but capital adequacy is one of the most important factors for each of these rating agencies.
Effective in 2014, A.M. Best Company began incorporating a type of economic capital modeling into the rating of all property and casualty companies by updating components of its property and casualty Best’s Capital Adequacy Ratio (BCAR). The objective is to incorporate a company-specific risk profile into the calculation of the BCAR score, as well as to tie insurers’ probability of default into the determination of the capital required to support individual rating levels.
Huggins Can Help with Economic Capital Modeling
Huggins employs industry-leading professionals with extensive knowledge and experience in economic capital modeling. Huggins has licensed Guy Carpenter’s MetaRisk© economic capital modeling software, one of the industry’s most transparent risk and capital decision tools. MetaRisk© gives insurers the power to see, understand and interact with the drivers of risk. Unlike the black box alternatives, Huggins’ licensed MetaRisk© model delivers clearer and deeper insights so our clients can make business-critical decisions with confidence. By using MetaRisk©, Huggins can provide over 180 customizable reports that can be tailored to the needs of each organization.
The Huggins MetaRisk© model is capable of using a large number of distributions for both the number of claims (frequency) and the average claim size (severity). Distributions available for modeling claim frequency include Poisson, Negative Binomial and Binomial, among others. Individual claim severity can be estimated within the model using statistical distributions such as Normal, Lognormal, Weibull, Gamma, Exponential, Generalized Pareto and Uniform. Most of these loss distributions are also available for use in modeling aggregate claim costs rather than individual claims. On an aggregate basis, the model can also include the use of copulas to incorporate the correlation between lines of business that can occur during catastrophes for instance. Copulas available for selection include Normal, Student’s T, HRT and Partial Perfect. Each of these copulas uses a different weighting system to account for the correlation between lines of business in the model, giving more or less weight to values in the tail of the distribution and varying the degree of correlation between each pair of lines of business.
Huggins also uses economic scenarios from Conning’s GEMS® Economic Scenario Generator (ESG). GEMS® is a state-of-the-art stochastic ESG that simulates future states of the global economy and capital markets using leading edge economic models, providing full market risk and asset class coverage. To estimate inflationary changes, both wage and the consumer price index (CPI) are projected based on econometric analyses. Estimates of investment returns and default risk are measured across a variety of investment possibilities including:
- US Treasury bonds
- US, United Kingdom, and Euro stock markets
- Emerging Markets stocks
- Blue Chip Stocks
- Master Limited Partnerships
- Real Estate Investment Trusts (REITs)
- Mortgage Backed Securities
- Corporate and Municipal bonds of varying quality
Data needed to Run the Model
In order to perform an economic capital modeling analysis, several types of data are needed. Balance Sheet information include assets (cash, bonds, common stock, other asset classes, surplus) and liabilities (loss and loss adjustment expense reserves by line or sub-line and payment patterns for existing reserves, unearned premium reserve, and other liabilities). Line of business inputs include direct written premium, underwriting expenses, earnings patterns, and claim counts. Reinsurance data include reinsurance contract terms along with specific catastrophe reinsurance terms, reinsurance catastrophe modeling results, ceded premium, ceded reinsurance attachment point, and ceded reinsurance limit.
Huggins’ Economic Capital Model Report Output
The Huggins ECM will produce pro forma financial statements including a balance sheet and income statement for a number of future years. As part of the ECM process, the model will calculate and display various financial values at a full range of confidence levels. The values at the confidence levels can compare results based on differing assumptions and include the effect of catastrophe losses. The output can also include value at risk (VaR) and tail value at risk (TVaR) for a number of projected financial values. VaR can be defined as the maximum potential amount that a value of a financial statement item “will experience within a specific confidence level over a specific period of time.” TVaR is a risk measure associated with the more general value at risk. It quantifies the expected value of the financial item given that an event outside a given confidence level has occurred. The TVaR is a measure of the expectation only in the tail of the distribution.
1 – Risk Management and Own Risk and Solvency Assessment Model Act (#505)
2 – James Lam, Enterprise Risk Management: From Incentives to Controls, John Wiley & Sons, 2003