Experience from Basel II Implementation and Suggestions for Improvement
As Basel II is implemented in the financial markets, we are receiving comments from practitioners about its methodology, rather than from researchers on the expectation of its suitability. Practitioners have put off Basel II implementation for as long as possible. You can hardly blame them; implementation is painful and costly. They spent seven years trying to water down the regulations before being forced to start using the new regulations.
Financial firms may have done themselves a disservice by not going with implementation in 1999-2000. In those days, over-the-counter (OTC) derivatives and structured products were in their infancy, while interest rate swaps (IRS) were considered complex instruments and required a great deal of negotiation to complete a deal. Today, you can trade, settle and clear an IRS electronically without much intervention from front or back office.
The work done by the International Swaps and Derivatives Association (ISDA) to bring standards to the OTC derivatives market has made it easier to trade instruments. The follow on work by DTCC to clear and warehouse trades electronically has reduced back office processing. It is easy to see that the standards and reduced back office workloads, coupled with potential profit, would lead to an explosion of new OTC derivative and structured product types.
We now have a situation where there are new types of trading strategies and new structured trade types created each month. Over the past year, we have worked with clients that are building new structures to reduce risk and create alpha. These are not firms that you would normally associate with being at the forefront of the market.
Standards and streamlined operations have lowered the bar to entry for the exotic derivatives market. Ready-made confirmations are available from ISDA or any number of the law firms specializing in the derivatives market. Small hedge funds or small portions of pension funds can trade complex instruments without hiring an army of specialized staff.
As usual, the market has outpaced regulation. It has very much outpaced the Basel II regulations because the industry fought its adoption for seven years. The regulations were finally adopted because they identify a real problem – measuring the risk from all operations. As firms adopt the regulations, they find that the methods and levels in Basel II do not always make a good fit with their current operations. I believe this is due to the delay in implementation.
Basel II was modified somewhat over the seven years of delay. The modifications were mostly regarding the redefinition or changing of levels within the standards. The discussion centered on the draft created circa 1999. There was little discussion about modifying the regulations to keep pace with the developments in financial engineering.
Now firms are applying the regulations to their portfolio and they are finding that it is difficult to express things in the way that is required by Basel II, which also gives them a fair representation of their true risk levels at the same time. If an earlier version of Basel II had been implemented in 2000 or 2001 this discussion could have instead been on the modification to handle new instruments and trading strategies.
I was not at any of the discussions to set the levels at Basel II. I do, however, hear a number of stories where each country suggested risk numbers that matched the actual levels in their respective countries. The final regulations, so the story goes, were a negotiated number among the countries. With the number negotiated, a model had to be reverse engineered to create a methodology.
You cannot blame country regulators for negotiating something that replicates what they already do. That was their experience and supported their belief that their system was the most prudent.
Because Basel II has been widely discussed and announced, it has raised the profile of risk management at financial firms. Basel II risk levels and methodology are now being used in firms outside the initial target of banks. We have clients in pension funds, insurance companies and hedge funds that use the methodology from Basel II. Together, we have questioned whether Basel II regulations are suitable for this task. It is, however, our only global set of standards for risk management. By that fact alone, it will receive widespread adoption.
Regulators have noticed some of these shortcomings and have highlighted the need for stress testing and measuring counterparty risk. As more and varied market participants trade complex instruments, it is important to measure the risk from counterparty default as well as instrument default. At counterparty default there will possibly be a default of numerous instruments.
Counterparty risk should be included into the normal risk calculations as perhaps a simulated exposure for each counterparty. The net exposure can be modified by existing netting agreements and collateral arrangements. Changes in counterparty credit rating, for a rated firm, can be simulated in the same manner as an exposure.
This is a simplified method to include such counterparty risks. Its advantage is that it is a unified approach and is transparent while capturing credit risk, netting and collateral. Most banks have internal rating systems that can create better ratings and transition data for the smaller unrated firms. This would be similar to the use of pools of asset-backed securities such as credit card debt. The characteristics of the pools would be used to rate an individual firm.
The use of negotiated levels does not provide the most robust regulatory tools. The call to use stress testing is beneficial to financial firms. Targeting risk to a single number is not necessarily a prudent idea. The market, counterparties, business cycles are all evolving. Some move together while others are not correlated.
A portfolio can appear well within the bounds of its risk limits based on current conditions. The strategy underlying that portfolio can have a fatal weakness that might be exposed under different conditions. Using stress testing to examine the portfolio under a number of different scenarios, such as changing interest rate spreads, forced downgrading of credit rating, downgraded business conditions and other factors can show the weaknesses in the portfolio. Stress testing will help find problems with the overall strategy as well as show the risk values under different conditions.
Over the past few years we have had a tranquil market. Interest rate spreads are tight, the equity markets have performed well and defaults are at an all time low. There is the danger of assuming that conditions will remain as they are indefinitely. Stress testing will permit you to see how your portfolio would perform in a less sanguine environment.
Another item we have discussed with our clients is the importance of correlation to the risk calculations. Depending on the method and length of data used to compute the standard deviation, significantly different correlation numbers are constructed. Basel II is silent about the methods used to calculate the risk numbers. Using different lengths of historical data can lead to different computed risk levels.
It would be useful if the Basel II working groups would take up some research on the underlying computations of the risk values. Transparency of the numbers would be beneficial to all market participants.
Using a more detailed factor structure can better elucidate the risk numbers. It would also be useful if the Basel II working parties looked at different industry and country factors. We have done extensive research on factor structure and believe that this area would also benefit from transparency.
Basel II has succeeded in highlighting the need to examine risk in financial firms – not just banks. It is a shame that the past seven years have been spent on negotiating definitions and levels within the regulations. Now that the regulations are being implemented in earnest, we have a group of skilled practitioners that have real world data which could make the regulations better serve their initial goal.
Should we return to the Basel II committee to create better computational methods to compute risk numbers or should we work outside the committee in trade and industry groups such as ISDA? Basel II is being applied outside its original target with many instruments that were not available at the time. The regulation has highlighted the need for good risk measurement. Now we need to improve the methods by which risk is measured.