Risk Models: Managing Chaos

Tuesday, September 13, 2011 ,

By Sinan Baskan

Like a doctor who smokes, sometimes people have trouble taking their own best advice.

For decades, financial services companies have been advising clients and fund managers to use diversification to mitigate risks and generate value. Portfolios well balanced across regions and assets with quantitative models generally perform much better over time with minimal risk.

However, when it comes to allocating their own capital to proprietary trading, financial institutions have not always done a good job of following their own advice. Note the state of traded mortgage-backed and credit derivative portfolios of leading institutions today.

Most market risk models used at trading desks are generally built around specific asset classes and do not take into account dynamics of the whole multi-asset portfolio. These models consider risks associated only with specific instruments, such as foreign exchanges or bonds.

The basic practices in risk management are ready for an update. It is time to push cross-asset risk monitoring up the organization.

Reengineering and Process

The fact that risk management practices need refinement will come as a surprise to no one. Eugene Ludwig, CEO of Promontory Financial Group in Washington, D.C., recently stated that financial services companies indicate a willingness to invest in modernizing processes and technology in spite of less profitable times.

In fact, the larger financial institutions have audited their internal risk processes and information systems, hoping to identify -- and, eventually, standardize and deploy - those that offer the best transparency and risk mitigation.

These processes and systems will form the basis of any future reengineering initiative, and they will set the bar for the industry as smaller firms rush to follow suit.

Reengineering the risk management function is not only about choosing the best processes. Indeed, when it comes to risk modeling and mitigation, two basic conversations occur: How are we measuring its effectiveness? And who has ownership of it?

Given the role that oversight (or lack thereof) played in the global credit crisis, it should be apparent that the second question deserves as much attention as the first.

The recent trend is for organizations to pass the ownership of risk monitoring and mitigation down to line-of-business managers. But their view of risk is fractured at best. Each manager can look at only a slice of the risk picture, and managers are sure to be influenced by the individual P&L requirements of their departments. These managers need tools and processes to understand risk distribution and mitigation in the context of a broader portfolio, so they can manage their own risk taking.

The Enterprise View

To address this need, firms must facilitate a heightened focus on risk management at the board level and downstream where trading takes place. This board-level oversight on governance, risk and compliance likely will take the form of a committee similar in mandate to those that have been established recently on compensation and audit.

Committee or no, organizations must come to grips with the fact that executive management has traditionally been too far removed from the discussion of risk. The critical factor in the solution will be providing top-level decision-makers with access to granular, transactional data from across the organization and all asset classes - securities trading, loan approvals and so forth.

These data will compose a highly accurate risk profile that can be built up and kept current to the balance sheet at both the P&L and enterprise levels. The balance sheet exposure at both levels must be transparent, current and recomputable on demand.

The importance of the enterprise view cannot be understated. Here, the data would facilitate decisions, such as instruments to which an institution might commit traded capital and in  what allocation ranges, as well as the limits for accepting additional risk in different categories.

The combination of top-down, end-to-end visibility and actionable authority at the board level can provide the organization with the best possible protection against calamity and, over time, restore consumer and shareholder confidence.

Compliance, Stability, Confidence

None of this will happen quickly. Although the incentive to change is present, organizations could take two to five years to make significant progress. Reengineering software systems and processes, then training staff to use them well, is a rigorous process.

The early signs of change will be incremental improvements in method and procedure. However, these early changes and the quality of the implementations and enforcement will foretell much about the long-term transformation that risk reform will achieve.

The immediate goals are to reach compliance with major regulations that are already in force, including the U.S. Dodd-Frank Act, Europe's Markets in Financial Instruments Directive and the global Basel III standards. Another, of course, is to deliver highly credible balance sheets that will encourage ratings agencies to evaluate them favorably.

There is also a more elusive goal: undoing the damage of the last few years and restoring investor and consumer confidence. Just saying that reform is under way will not be enough. The public will not be comforted by recondite promises of aggregated real-time risk data. If the industry commits to a top-down, principles-driven approach that verifiably results in market stability, then investor confidence will eventually follow.

As director of business development for financial markets at Sybase, Sinan Baskan (sinan.baskan@sybase.com ) is responsible for developing solutions for lines of business in the financial services sector. He has held various positions in the product engineering, professional services and marketing organizations at Sybase. Previously, he worked at HSBC Corporate Investment Bank on risk analytics.