Evolving from Quantitative Risk Management to a High-Performance Risk Management Analytic Framework

WHITE PAPER Evolving from Quantitative Risk Management to a High-Performance Risk Management Analytic Framework Insights on a new direction for risk management by Myron S. Scholes, PhD and Nobel Laureate, and Tom Kimner, Head of Americas Risk Practice at SAS EVOLVING TO A HIGH-PERFORMANCE RISK MANAGEMENT ANALYTIC FRAMEWORK Table of Contents The Evolution of Risk Management........................................................ 1 Moving Toward Optimization................................................................. 4 Quantitative Math or Management Analytics?......................................... 5 Shifting from a Portfolio-Theory View to an Optimization View of Risk Management....................................................................................... 7 Risk Management Is Capital Optimization............................................... 9 Capital – Diversification or Optimization?............................................. 10 Implementation and Management of a High-Performance Risk Process... 12 Implementation of a High-Performance Computational Environment....... 14 Tangible Business Value of a High-Performance Risk Process................. 15 i EVOLVING TO A HIGH-PERFORMANCE RISK MANAGEMENT ANALYTIC FRAMEWORK The content providers for this paper were Myron S. Scholes, PhD and Nobel Laureate, and Tom Kimner, Head of Americas Risk Practice at SAS. ii EVOLVING TO A HIGH-PERFORMANCE RISK MANAGEMENT ANALYTIC FRAMEWORK The Evolution of Risk Management As a result of the recent market shocks, banks, capital markets firms and asset managers are rethinking certain issues and focusing on: 1) How to integrate not only risk and reward tradeoffs using portfolio theory, but also how to plan for market shocks; 2) The resulting impact of these shocks on the business and its divisions. Leading financial entities are linking their portfolio risk with the return on capital and integrating market liquidity into their analyses in an attempt to gain a more complete view of risk and return. As a result, optimization of capital deployed – rather than just a single view of risk exposures – has become the new role of risk management. Risk management has traditionally relied on expert judgments coupled with a ■ Optimization of capital deployed – narrow use of quantitative techniques. These techniques are being replaced rather than just a single view of risk by sophisticated, but broadly used analytics that make traditional quantitative exposures – has become the new techniques more transparent and available to decision makers by combining the role of risk management. quantitative techniques with an analytic decision framework that is optimized for exposures and capital return. The analytics of decision science are combined with quantitative risk-exposure calculations within a dynamic option framework, which merges the fragmented elements of risk and return decision making. Predictive, on-demand scenarios are the core business tools that leading financial firms will adopt in the coming years. Without an up-to-the minute, scenario-optimized view of risk and return, executives are unable to understand or integrate the performance of capital deployed in various asset classes or divisions of the firm. By incorporating all elements of the risk and reward equation – exposures, return, capital reserves, capital deployed in various forms, firm liquidity and market liquidity – we now have the opportunity to provide and grow high-performance risk-management capabilities within firms. Portfolio theory has dominated the thinking of capital-management strategy planning, and value at risk (VaR) has dominated the quantification of risk management exposures within the portfolio theory construct. Traditionally, accounting and planning for liquidity shocks and market dislocations has been handled indirectly through techniques such as scenario analyses or forms of risk budgeting. However, scenarios were typically only constructed to deal with interest rate shocks or a change in top-level economic indicators. Although common risk techniques such as VaR and probability of default are still employed, they fail to anticipate systemic changes in the structure of markets. These techniques assume that the volatility of the market and correlations among assets change slowly or not at all; they are not designed to handle systemic negative changes caused by jumps in the availability of liquidity or jumps in market values. 1 EVOLVING TO A HIGH-PERFORMANCE RISK MANAGEMENT ANALYTIC FRAMEWORK As we emerge from the crisis that began in 2008, not only have banks had to boost capital reserves, but the impact of sovereign nation debt has also constricted the flow of capital. This has the potential to heighten the impact of market volatility, and with unanticipated events, we are anticipating that there potentially will be an amplification effect of the volatility and market shocks. The ability to acquire capital for investment or to liquidate a position potentially will accelerate more default events over the next few years as markets adjust to systemic changes in the market structure. Although most firms use dynamic measures such as VaR to gauge the sensitivity ■ Although most firms use dynamic of results to short-run market factor movements, they realize that they need to measures such as VaR to gauge overlay these measures with additional capital and static reserves to handle shocks. the sensitivity of results to short- Correctly, entities realize that short-term measures are inadequate. However, new run market factor movements, they work is needed to measure the size of needed risk reserves or cushions; that is, how to dynamically adjust them and partition the cushion among the various asset realize that they need to overlay categories within an entity to make more accurate risk/return tradeoffs. This is a new these measures with additional direction for research. The ability to enhance risk methodologies is due to advances capital and static reserves to handle in technology, such as the SAS® high-performance computational environment, shocks. that remove the computational complexity associated with multifactor, cross-firm, full-valuation methods. Large regression computations of thousands of risk factors, across thousands of market states, can be computed in minutes rather than days. Breakthroughs in methodology can now be realized due to breakthroughs in technology. The three tools of risk management are diversification, reserves and insurance. Each tool has its benefits and costs. For example, financial entities concentrate their holdings to earn returns because they need to service client demands – thereby sacrificing diversification. Too large a reserve held in the form of low-yielding bonds (including low levels of debt) protects against shocks, but limits investments and expected returns. Additionally, insurance is costly if hedging needs and timing are unknown, and the “correct” form and amount of insurance is unknowable. For example, most of the time, financing activities with overnight wholesale funds is less expensive than term financing until term funding would have saved the entity from collapse. Moreover, each tool might be insufficient at times of market shocks or systemic risk. Moving from theory to implementation issues, market participants relied too heavily on recent market experience (during the 1990s and 2000s) to frame their views on risk and to calibrate their models. They concluded incorrectly that the likely need and the resultant cost to adjust their holdings – and to reduce risks in light of shocks and lack of liquidity in the market – were extremely low. They relied almost exclusively on the advantages of diversification across uncorrelated firm activities and concluded that risks were controlled within the isolated portfolios; they relied too heavily on a limited set of quantitative techniques to measure and to plan on how to react to unexpected market conditions. 2 EVOLVING TO A HIGH-PERFORMANCE RISK MANAGEMENT ANALYTIC FRAMEWORK They also relied extensively on external monitors, such as the rating agencies, to validate risks. The rating agencies failed to incorporate multiple, simultaneous failures in their models; they also overlooked the fact that recent market event data might not tell the complete story, or that the quality of the composition of structured products might deteriorate over time as entities reverse-engineered them to “just pass” to receive a rating of “AAA.” Therefore, the problem is that firms simply maximized along a truncated view of possible investment paths, assuming that recent volatilities and observed ■ The problem is that firms simply correlations were the best indicators of future volatility and correlations.