RISK EXECUTIVE SERIES
James J. Jockle, Chief Marketing Officer at Numerix, discusses the latest challenges and opportunities involved with implementing an effective model risk framework.
When it comes down to managing model risk, what are banks doing these days? What is the buy-side doing? What’s missing, and where is all of this heading?
As discussed by Numerix CMO, Jim Jockle at SPA-2015 - The 11th Annual Structured Products Association’s Spring Conference recently held in New York City, the challenges in dealing with model risk continue to be on the forefront of every derivative practitioner’s mind.
In his SPA presentation, Mr. Jockle explored why model risk is important, along with highlighting the three main drivers behind today’s model risk challenges, including:
- Regulatory Pressure: One of the most persistent and important drivers for model risk management across financial services is regulatory pressure
- Stakeholder Pressure: Stakeholders are playing closer attention to the process of managing risk, especially the use of risk models
- Managing Reputational Risk: Firms realize that model failures could cause significant reputational damage and want to be able to include reputation as part of model risk assessment.
The Road to Model Risk Management Independence
While the concept of model risk has been around for many years, the SR 11-7 Guidance on Model Risk Management, as issued by the Federal Reserve Board in 2011, continues to be one of the most important pieces of legislation when it comes down to model risk—despite the fact that is only five pages in length and leaves room for interpretation. Overall, the Fed guidance is a principle-based document—not prescriptive. Basically, SR 11-7 defines a model as a set of calculations and defined inputs used to generate a recommendation. In the case of structured products, the model output is a valuation or risk metric.
As many of us are familiar with by now, model types include but are not limited to: pricing; trading; counterparty risk; enterprise risk; and most importantly in the context of Structured Products: “a pay-off + a stochastic process.” The Fed Guidance calls for the adaptation of a completely independent model risk management process. Looking forward, research indicates that many banks are looking to have this fully instituted by January 2016.
Model Risk Framework
From model development and validation, to documentation, inventory, approvals and ongoing governance—the process takes time. Research indicates that a firm’s model risk framework processes can take anywhere from 3 weeks to 18 months based on complexity, when handled manually.
So, Why Does this Matter?
The length and complexities of the model risk management process can greatly impede a firms’ time-to-market. Moreover, constraints on PnL as it relates to these trades must be considered. Some firms speed things up, fast-tracking the models through the use of model-based limits. However, research shows that other institutions are pulling away from the use of limits and waiting until the models have been fully validated and tested.
During his SPA presentation, Mr. Jockle also engaged in a discussion on the trend emerging toward the use of projections of future performance in Europe—though the use of future projections has not yet been accelerated in the US. Looking forward, many regulatory authorities are seeing the value of projections via stress testing and scenario generation as part of the banking book. However, it is still to be determined how forward projections will be ultimately used in this market.
Lack of Automation and IT Spending
Research shows that many buy-side firms have heeded the call of SR 11-7 back in 2011 and feel somewhat more confident in their model risk framework processing overall. However, a recently released analyst report “Reducing the Risk of Using Financial Models,” authored by Kevin McPartland, Head of Market Structure and Technology Research at Greenwich Associates, reveals the somewhat surprising lack of automation and IT spending amongst sell-side banks when it comes down to Model Risk Management. Specifically, the report cites, how the “…Lack of reporting standards leaves firms spending unneeded hours demonstrating to regulators that they are complying with the spirit of the rules” set forth in the Guidance.
And while recent Greenwich Associates research found that both sell side and buy side firms have improved their model risk management practices in the past few years, those process remain largely manual, very time consuming, and prone to disputes between different departments with differing objectives and incentives.
Room for Both Improvement and Opportunity: Leveraging Technology
The Greenwich Associates report concludes that the movement by financial institutions and regulators to develop standardized processes and technology for managing financial model risk would lower costs and potentially reduce systemic risk.
In his SPA presentation, Mr. Jockle reiterated these sentiments, also adding that moving beyond the checkbox and investing in automated processes would enable these firms to capitalize on market opportunity, and enhance efficiency with faster time-to-market. In addition, through the use of automation, firms can achieve better risk identification with heightened granularity.
For a more detailed analysis, view our on-demand webinar, "Challenges & Opportunities in Dealing with Model Risk: A Panel Discussion with Greenwich Associates & Numerix."
