Introduction to Predyct-ERM continued

ERM is a Single Period VaR Model

One of the principal issues to understand about ERM is that it is a VaR-based single period simulation model, although it produces an entire family of VaR measures including Tail VaR (TVaR)—also known as Expected Shortfall (ES), Marginal VaR, and Incremental VaR (IVaR). VaR is defined as the worst loss that a company may experience over a target period (one year) with a given level of confidence (see Appendix D.1 for more details). If a company’s VaR is $100 million at the 95th level of confidence, this means that there is a 5% chance of losing more than $100 million of net worth over this period of time. VaR is always assessed at what is called the horizon and the horizon period for ERM is one year.

ERM is a Cash Flow Model

ERM is a cash flow based model that marks all financial investments to market. ERM marks insurance liabilities to model, i.e., the value of the liabilities is the expected value of their future payments over the life of the obligation present valued to the horizon. Earnings in ERM are defined as changes in the value of net worth (assets minus liabilities). For investments, this is not dissimilar to statutory accounting principles where such factors as realized and unrealized gains/losses are either added to or netted from a firm’s surplus. The difference, however, is that statutory accounting principles do not run these credits and debits through the income statement. ERM, on the other hand, converts the firm’s income statement from accounting values to mark-to-market values so that realistic rates of return on risk adjusted capital (RAROC) can be attributed in the current accounting period.

ERM Produces a Multidimensional Picture of Risk and Risk-Adjusted Performance

ERM is functionally all about producing two kinds of values:

Risk measures—the company’s total risk, the risk contribution of individual business segments (on both stand-alone and allocated basis), and contribution of individual risk categories (insurance, credit, interest rate, equity, and forex risks).
Performance measures, with the focus on valuation of cash flows that go toward the measurement of the firm’s risk adjusted return on capital (RAROC).

Contributing to a firm’s risk will be the principal risk categories to which non-life insurance companies are exposed: Reserve Risk (old business), Underwriting Risk (new business), Equity Risk, Interest Rate Risk, Credit Risk, Foreign Exchange Risk, and Catastrophe Risk. One can envision that the company’s VaR will be influenced over the next year by each one of these risk categories. To understand how this works, one has to consider the cash inflows and cash outflows of an insurance enterprise:

Risk Categories specific to Insurance

Underwriting Risk is the risk associated with the new business that will be written over the target period (between the evaluation date and horizon). This risk has two sources: the uncertainty of the new premiums that will be collected over the period and the uncertainty of the future loss payments on the new policies. Income that is allocated to this risk category for RAROC purposes is represented by the present value of the collected premium, minus the present value of the expected loss and expense, plus interest on the Economic Capital allocated to support the Underwriting Risk (see Sections 3.2.1.5 and 3.4).
Reserve Risk: Business that was written in the past that still remains on the books and for which the company has outstanding reserves is the source of Reserve Risk. Reserve risk is defined as uncertainty of the loss payments associated with the prior accident years. Income allocated to the Reserve Risk for RAROC purposes is the present value of duration matched interest on the insurance reserves plus interest on the Economic Capital that is allocated to support the Reserve Risk (see Sections 3.2.1.5 and 3.4).

General Risk Categories

Interest Rate Risk: Cash flows generated by ERM are discounted in accordance with the term structure of interest rates. ERM simulates individual interest rates for a specified set of maturities and captures both parallel and non-parallel shifts in the yield curve. Interest rate risk is incurred when there is a mismatch between the company’s assets and liabilities. The mismatched income is subject to the interest rate risk of the specific time bucket in which it occurs. Cash flows are assigned to time buckets (vertices) in accordance with the RiskMetrics methodology that is covered in the next section.
Credit risk is defined as uncertainty of value due to changes in credit quality. Credit risk in ERM is measured using the CreditMetrics approach which is based on the extended Merton model. Merton’s insight was to recognize that the equity in a firm can be considered as a call option on the firm’s assets. As a result, the value of the firm’s debt can be expressed as the amount of liabilities outstanding reduced by the put option on the assets (the strike being the amount of liabilities). In this option theoretic valuation of debt, bonds become riskier if the return on the underlying equity is weakening, if the maturity is long versus short, if the volatility of the equity is higher rather than lower. The put option reduces the value of debt due to the possibility of default. This basic Merton model can be easily extended to include rating changes (see Section 3.2.1.4).
Equity Risk: Equity returns are sensitive to systematic and idiosyncratic factors, all of which are captured in ERM. Systematic factors include sector and country returns. ERM covers 10 equity sectors in all developed countries and many developing countries.
Foreign Exchange Risk: Income from foreign operations is accounted for in the company’s home currency. Foreign income is subject to the same risk factors as income produced in the home currency, plus forex risk.
Catastrophe Risk: Losses due to natural catastrophes may have a high impact on the company operations. CAT losses are rare events with high uncertainty which contributes to the overall insurance risk. Also, the large reinsurance receivables generated by the ceded CAT losses are usually the largest single contributor to the credit risk of an insurance company. When aggregating the catastrophe risk with that of the regular losses within a simulation model, the challenge is to properly estimate the impact of the rare CAT events on the overall VaR of the company (see Appendix D.3).

The amount of equity capital that a firm must ultimately carry to support all these risks will depend on its VaR. A large VaR will signal an advanced warning to the firm’s leadership that it could potentially lose a level of capital that may impair its operations in the eyes of its stakeholders. This risk is contributed to the VaR from the principal risk categories that have been discussed. But this risk is also mitigated by the degree to which these risk categories are interconnected. For most companies, a significant reduction in the required risk capital (between one-quarter and one-half) could be attributed to correlations inherent in the company’s assets and liabilities. So it is extremely important that firms be able to compute diversification benefits with as much accuracy as possible. Issues of correlation are discussed throughout most sections of this paper (especially see section 3.2.3 and Appendix B:).

Virtues of Predyct ERM

The principal virtues that we identify in favor of ERM are:

ERM is an integrated model that measures all risks and their interdependencies within a single framework
All assets are processed at the Committee on Uniform Securities Identification Procedures (CUSIP) level and consistently marked-to-market
All liabilities are marked-to-model and valued on the discounted cash flow basis
Earnings are defined in terms of net worth (assets minus liabilities) rather than in accounting terms
The capital that is required to support the company’s risk (and individual business segment risk) is identified at a given level of confidence
Capital is allocated to support each business segment
The earnings performance of each business segment is computed to reflect risk-adjusted returns on allocated capital rather than accounting-based measures of earning performance that are inherently flawed.

Hopefully, this description will assist the reader by providing a “big picture” of ERM as subsequent sections delve into substantial detail of how its capabilities are actually executed.