With the credit crunch now established as a global phenomenon and financial markets in disarray questions are being asked about the capability of risk algorithms used within banking systems supposed to provide layers of protection for the financial services firms and investors. The very fact the financial markets are in this disastrous situation and banks once revered as pillars of the global financial system have had to go ‘cap in hand’ for bailouts, demonstrates that the risk systems and the banks ability to use them effectively failed. How could this happen and what are the ramifications for the future?

All risk measurements start from a basis of historical facts and these form the benchmarks that similar situations are marked against. The markets and economies all tend to go in cycles with only the length of each cycle undetermined. However, chartists would say that by cataloguing data over years the historical trends can be first measured and then predicted. The problem is that with acts of God or unintended consequences, any predictive path can be thrown into a quandary.

Already the current financial crisis has been recorded and included within the historical data of risk systems but it is as yet too soon to produce any definite conclusions to assist in decision making. But it is only going to be a matter of time before we begin to evaluate the risks in the current market to build new risk scenarios and models to test future investment.

The problem with this historical data approach is that it is based upon past events rather than predicted scenarios and ‘what if’ factors and therefore can produce knee jerk reactions within the investment industry, as new price levels create new buy or sell orders. The result can produce extreme volatility within the equity markets as they move violently up or down on relatively small amounts of actual trade volume. OpenLink is one of the most respected software firms in the market and have been established long enough to see many extreme market conditions (although probably not like today’s) how do they see the impacts of the financial crisis on risk systems?

Ken Knowles, executive vice president, OpenLink Financial & Risk Solutions, comments: "The monitoring of risk ultimately relies upon on a clear understanding of the exposures, how they are changing, what is making money and why, and finally, what could go wrong.  Risk systems need to have breadth, flexibility and depth of functionality in order to operate within the volatile markets we are seeing.  Asset classes are increasingly converging, and the ability to view your entire market position and associated risks across those markets will continue to differentiate the successful organizations in today’s tough market environment.  What is required is a product that spans all asset classes and is fully integrated with middle and back office functionality – from front office risk management and profit and loss reporting to confirmation, settlement, reconciliation, payment and accounting.  Increasingly, vendors are offering this complete front-to-back office functionality for cross-asset trading.  It has become a reality.

"Computational complexity plays a significant role – if it takes 20 hours to run risk on a complex derivatives book, then those same approaches certainly can’t be used to manage risk in a real time trading environment.  Some tradeoffs and compromises must be made, and unfortunately we tend to hear in hindsight about the approaches that didn’t work so well.  There are also additional factors to take into consideration – for instance the information lag from the time numbers are computed to the time that information is then distributed through the organization.  Liquidity risk is another issue coming to the fore, as extreme events and market conditions may trigger the downward liquidity spiral as the access to capital dries up."

It is clear that banks have overestimated the adequacy of their risk capabilities but is it a weakness only found in their risk systems or flawed management with people responsible, unable to fulfil their functions? The combination of these weaknesses would certainly have been enough to open up a risk black hole, sucking the banks in before they could find remedies.

It used be a simple banking decision to lend against the ability to repay and the quality of the collateral offered to protect the banks. But the increasing complexity of derivatives and the lack of transparency as more and more were created in the OTC markets somehow bypassed the risk systems used by banks and began the march towards the credit crunch and then the more widely felt market meltdown and economic and social failures.

Banks should also have been more aware of the risk of collusion between departments with people gaining from breaches of risk controls. The management structures, which should provide a multiple process of authorisations of trades appear to have proved ineffective and together with the other flaws in the risk management system failed to provide strong enough barriers to prevent the scale of the financial crisis.

So on the surface the breakdown of the banks risk protection barrier was a combination of inherent flaws that created the inability of financial institutions and their regulators to detect stresses within the banks and the markets. However, if these breakdowns were not bad enough they were compounded by the underlying links with the economy, which had been caused by government policies. This then became a powerful force that prevented any individual bank or government from taking unilateral measures to mitigate the impending disaster.

The financial hurricane was now unstoppable and with liquidity within markets drying up, the preferred government solution was to buy stakes in the banks and pump billions into the markets in a massive attempt to slow down, stop and then reverse the global economies. The inability of all banks to manage their own risks within an overall banking industry measurement barometer has been culpable in the creation of not only the immediate crisis but then its escalation into the global economy.

Hindsight is a wonderful thing and it can be used to good effect in this crisis. What is required is the development of predictive risk models that can measure more acutely the global shifts in the financial markets. There is an urgent need to design better risk systems to capture the complexities of financial products within the overall industry mechanism, with the stresses highlighted in both domestic and cross border business. The central banks need to be brought into the risk detection system so they can bring potential problems to the attention of the banks, regulators and governments.

Ken Knowles, executive vice president, OpenLink Financial & Risk Solutions, comments: "In a year where the topic of risk has rarely been out of the headlines, financial institutions have been really forced to sit up and take notice of the systems and policies they have in place.  Given the current state of the market the industry has seen that no one is above reproach and even the mighty can fall.  Whilst most financial institutions do place risk at the forefront of their thoughts, the key is to now examine and understand why the systems failed at a time when they were needed most – was it a mathematical issue, a computational complexity issue, or a business modelling issue? 

"The tough market environment, coupled together with the ever-increasing complexity of financial instruments, means that vendors are fundamental to the successful operation of the industry.  Accordingly, there is likely to be increased investment in risk systems and risk management as a whole, which is very encouraging as institutions, governments and regulators alike appreciate how integral it is to the stability of the market.  In the wake of the crisis, periods of uncertainty and reactive policy decisions are also likely to follow, which will force the incumbent risk systems to adapt to the changing requirements.  This will be a real test of the software infrastructure, and those with well-designed extensibility will be better equipped to manage the dynamics of these risk policy requirements and decisions."

As the finance industry recovers it must learn by this terrible experience and go back to the drawing board in creating a combined risk management mechanism that gives early warnings to the markets and the government of potential problems. A revaluation of risk models and to move towards predictive technologies would appear to be the obvious development and one that I am sure will be recognised by the software firms and their Gurus.   

By Gary Wright, MSI