It's more than just pipes
Like monetary policy, risk infrastructure has become a sexy topic. Three stories from large North American banks indicate why.
One institution sought to grow its business through increasing its volume of trades in credit instruments, equity derivatives, and foreign exchange. The regulator surprised them by throwing up a stop sign, saying no to the increased volume of business. Why? The regulator was not confident that the institution had the risk management models, processes, and infrastructure in place to safely trade higher volumes of these financial instruments.
Another institution had put out the fires of excessive risk and wanted to concentrate again on growth. Management developed a trading strategy that sought to balance return, capital, and risk. However, they found that they had insufficient real-time information to know which trades to make in the short term, or which markets to stay in for the medium term. Attempts to grow would be fruitless until they solved this problem.
A third institution had recently acquired one of their largest competitors with a very large business in the capital markets arena—much larger than their own. Overnight, this organization migrated from a small, predominately North American trading business, to one of the largest global trading businesses in the world. They quickly realized that they were not set up to manage the risk. Risk and regulatory reporting became slow and riddled with errors, which attracted unwelcome regulatory attention and oversight. Searching for a way out, their solution was to completely revamp everything that went on behind the scenes.
KPMG is involved in each of these stories and numerous similar situations. What are we learning?
Simply, many banks are hampered in their business objectives by inadequate risk infrastructure, and by this, we do not mean just technology, but equally data, analytics, workflow, organization, and reporting the total package.
Before the financial crisis, a dollar spent on risk management was often considered a dollar that could not be bonused to people or traded for profit. Risk management was more of an afterthought, focused on basic risk reporting, meeting minimum regulatory standards, keeping pace (or trying) with the business to bring a measure of control, all the while using systems and approaches that required huge amounts of data management, one-off spreadsheets, and tons of reconciliation, only to produce information that was either late to the table or only somewhat relevant. The financial crisis exposed this house of cards. Many banks found serious problems around risk, systems and information that they and their regulators are now becoming aggressive about solving. For example, many risks, such as liquidity risks, "fat tail" events, or credit in the structured products arena, are not reported against. Governance for them is either poorly defined or not defined at all. Systems being used grew up in "stovepipes" that do not communicate with one another and thus ignore the inter-connected nature of risks. There is data but not information; the "why," "so what," and decision implications are missing. Measurement rarely addresses success or failure. What is the benchmark? What is "good"? Risk management staff spend half their time or more on data production, reconciliation, and reporting—not on analysis and decision support.
And the solutions in the current marketplace are only somewhat helpful. The vendor marketplace is one serviced by multiple providers with overlapping and not consistent services. The right application or more typically the set of applications is dependent on each bank's product/service suite and its overall business objectives. To show the variety and the complexity, here is one overview of the systems solutions marketplace. It captures only some of the various solution providers.
Our experience teaches that the leading risk systems struggle to meet the clients' endto- end needs. The following risk management capabilities are often lacking in off-the-shelf solutions:
1. Limit breaches: leading market solutions are not necessarily designed as workflow engines:
a. The limits are often rigidly defined and may not be flexible enough to meet the unique needs of different trading businesses.
b. Intra-day reallocation of limits and limit breach exception processing are not enabled and often need to be custom built.
2. Risk reporting: packaged reporting capabilities often meet the minimum reporting requirements, but not management's needs for a more robust capability:
a. Applications do not allow the frequent access, exploration, and analysis of data
b. Additional reporting needs typically require customization
c. Reporting does not integrate risk, performance, and capital into a single information framework.
3. Measurement methods often no longer meet basic regulatory needs, as the approaches are devised to meet a broad range of needs, and not the specific complexity of any one given organization. Further, the emphasis is on broadly used analytics, rather than analytics devised for use to manage a specific universe of risk and financial characteristics.
4. Data is not managed in a way to support risk management work flow; rather, it is modelled and optimized to support a specific analytic framework, which means the focus is on the measurement of risk, not the management of risk.
The way forward
So what is the way forward? KPMG has recently interviewed both banking clients and non-clients to ascertain their needs and their requirements. We found their needs are straight-forward and in fact a bit "motherhood and apple pie." Banks want electronic delivery of forward-looking, robust, integrated measures and information that are timely, easy to analyze, and useful for what-if scenarios. Easy to say, but much harder to do. Their resulting must-have requirements for technology investments are the following:
- Robust technology is essential for risk managers to move into proactive, analytical roles, instead of being forced to focus only on reporting, reconciliation, and data management
- Systems need exception reporting and/ or management alerts to focus attention on problem areas
- Technology must apply a formal "process and policy execution" framework to trading floor activities
- Old stovepipe systems must be migrated to more holistic risk architectures
- Systems must ensure the quality, transparency, and traceability of data
- Information should be electronically produced and delivered.
Meeting these requirements demands first a holistic framework, such as the following:
The goal is improved risk management performance. Technology addresses the sources of data, its transportation, its transformation, and its storage. Equally important, we are stressing, are the work flow processes and organization structures, the choice of measures and analytics, and the reporting and presentation. Finally, what are too often ignored are the governance structures and processes that bind it all together. Failure to address clear accountabilities, the governance hierarchy, and transparent dispute and exception escalation procedures, always results in serious breakdown.
A Case Study
In one instance, a top five trading institution found that as the crisis unfolded, they did not have a good command of the risk being taken by the traders, whether they were being paid adequately for that risk, and how that risk might behave under various scenarios of importance to different management levels in the organization. Further, the risk management organization was predominately a producer of information, as opposed to an analytic support and active control function—an issue that management believed must be addressed. All of the basic regulatory minimum standards for risk management had been met, yet it was still not enough.
Our team was engaged to quickly identify and decompose the risk in the trading book, then frame a robust risk and performance management framework that provided actionable business insight and information with clear connectivity into risk/ regulatory reporting and capital. Once the work had been completed, we were asked to devise a practical strategy to implement what amounted to a fundamental revamp of the information used by risk management, executive management, and the trading organization. The revamp needed to leverage to the extent possible the legacy environment to enhance the speed of deployment while minimizing the cost to implement. The objectives were not to just implement new analytics, but to design a capability that shifted human focus from data management, reconcilement, and reporting, to analysis, investigation, advisory, and oversight.
One of the most interesting aspects of improvements to risk infrastructure is that many projects can be paid for with the savings they create. When the reconciliation and data management burden is cut in half, cost reductions of 10 to 25 percent can be achieved. Time for risk analysis and oversight can be increased by up to 250 percent in our experience, and the quality of oversight is enhanced by automation, management alerts, and new risk views.
Also, a better risk infrastructure doesn't just make for better risk managers; it can make for better front office performance too. Faster access to information and enhanced analytics can increase speed to deploy by 35 percent—with better control. Valuation can be 30 to 50 percent faster, with greater transparency around the migration of key assumptions, all while helping to reduce the work effort required to complete a reliable mark to focus on other essential elements of risk management.
Before the financial crisis, trading organizations let the front office limit the intrusion of risk infrastructure on their activities. Risk managers concentrated on reporting, reconciliation, and data management, and had little time for truly value-added activity in conjunction with their front office peers. Even the front office didn't realize they were sacrificing performance by avoiding these investments and not embracing enhanced risk management capabilities.
Now, after the financial crisis, regulators are keenly interested in risk infrastructure from the point of view of analytics, information, process, technology, and governance. Although the concepts involved are relatively simple, and the benefits tangible, many financial institutions are not yet moving quickly in this direction. Risk infrastructure is, realistically, a post-rebound project that requires longer-term strategy, planning, and execution. However, it is where regulators and the banks should go. The benefits should also make it a place where banks want to be.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.