Avoiding the Pitfalls of Choosing a Risk Management System

By Nigel Hooker, DFA Capital Management

With the financial crisis on Wall Street and the threat to the world economy, never has there been more focus on risk management practices in banks and insurance companies. It's now more critical than ever before to know how much capital is needed to support current and future business and which segments create shareholder value or destroy it.At the core of financial risk analysis are systems that enable companies to model possible future economic and financial scenarios to gain insight into the financial risks that could damage their firms. Choosing the right tools for risk modeling can make the difference between success and failure so it's imperative to consider the choices carefully to avoid common mistakes when establishing a risk management system.

Typical risk management systems include two main components: an economic scenario generator (ESG) - a tool that is used to model the economy, financial markets and other external influences - and a stochastic modeling tool, which applies the economic scenarios to the company's risk exposures to simulate the results of the financial decisions made, and evaluate performance and risk metrics based on the output. How these two components work, individually and together, critically affects the reliability of the model outputs and whether they truly support better risk decision making.

ESGs are composed of several types of models that work together to depict the economy and financial markets. Within the ESG, the most important elements are frequently its equity model and its interest rate model. Keep in mind that not all equity models are created equal. The best equity models incorporate stochastic volatility and include random shocks that make it possible to reproduce the kinds of extreme scenarios observed in the real equity markets. Banks and other companies are now learning that modeling such real-world behavior accurately is critical to assessing the risk in investment portfolios and uncovering weaknesses in hedging strategies.

Understanding how an ESG models interest rates is also essential to ensuring that models are arbitrage free and that derivative pricing remains reliable. The best ESGs provide a risk-free (government bond) term structure with real-world dynamics that reliably reflect past behavior while producing a plausible number of more extreme yet conceivable scenarios. The fall-out on Wall Street indicates that some companies were not using models based on sufficient historical data to include extreme events similar to what we are experiencing in the markets today.

An ESG should also be looked at as a whole. Is it comprehensive enough to properly model the correlations between different economies and currencies? As we have seen in recent weeks, the world's economies and currencies are intricately linked. Can you use the ESG to model a rich set of asset classes, including derivatives and other complex instruments? As recent history illustrates, the tools many companies were using to model investments had limitations that failed to expose the true risks of those investments.

Good ESGs model defaultable securities through both systemic and idiosyncratic components of risk, at security level, ensuring that simulations will reflect accumulation and concentration risk. What that means is that corporate bond models should be able to faithfully reproduce rating transition dynamics, default rates and the interaction between equity markets and credit spreads by rating class.

While the ESG models the economy and financial markets external to a company, these can only be applied to the company's risk exposures through a stochastic simulation tool. A very common mistake companies make is to choose a tool that is only capable of modeling part of the business. A whole company model supports aggregation of data within complex business structures. Even the best modeling tools will fail to achieve their goals if they do not provide visibility to risks at a whole company level, thereby misrepresenting the risks a firm is truly facing.

The best modeling tools for a bank can be applied across the full range of its business segments; they can store output at detailed level, allowing full drill-down analysis of results; and can present output in both market consistent (fair value) and accounting-based views, recognizing fungibility (or not) of capital across disparate entities and jurisdictions. We've recently seen how the inability of healthy business units to support ailing ones has contributed to the failure of some firms.

Although selecting the right modeling tools for risk management is essential, one further mistake companies commonly make doesn't have anything to do with tools. It is essential to ensure that corporate culture avoids the typical silo approach to running a business. As we continue to follow news on the economy, it becomes clear that companies that conduct risk management in business silos expose their firms to unnecessary and avoidable risks. Tying true enterprisewide risk management to business performance management, along with implementation of the right tools, is the only way for companies to ensure long-term success.

Nigel Hooker is EVP, Professional Services, Europe, for DFA Capital Management Inc., a provider of enterprise risk management software for the insurance and financial services industries.