Documentation Risk, Operational Risk and the New Capital Accord

Ladies and Gentlemen,

I am very pleased to be able to speak at this special luncheon to launch the APLMA Standard Documentation. I am told that there has been an unprecedented demand to attend this event. I would like to think that this reflects my own attractions as a speaker. But I am realistic enough to recognise that the real star of the show is the documentation itself; and your attendance here today reflects the level of interest in the outcome of the Association's hard work in bringing the first phase of this project to fruition.

My purpose today is to stress the importance of good documentation and to place it in the more general context of the increased regulatory focus on operational risk that is a feature of the Basel Committee's proposed New Capital Accord. I will also briefly describe the latest developments in the New Accord and bring you up to date on the revised timetable.

The work of the Association

It is over three years since I spoke at the inaugural luncheon of the Association in Hong Kong. I said on that occasion that the Association promised to become a vital part of the market infrastructure in Hong Kong. I noted particularly that the Association could play a role in three main areas: by acting as an industry representative and interlocutor with regulators; by establishing common standards and documentation; and by promoting market transparency and efficiency.

I am glad to say that the Association has delivered on its promises in each of these areas. Its website is, for example, now up and running, and is an important source of information for both members and non-members. If I could put in a plea, however, it would be for more market statistics to be made available on the parts of the website that are open to non-members - or honorary members such as the HKMA.

In its representational role, the Association offered very helpful comments last year on the draft guideline that the HKMA prepared on the management of the risks associated with syndicated lending. Our initial draft generated a certain amount of heat. This was I think largely due to a feeling on the part of the industry that we had overplayed the inherent risks of syndicated lending. We therefore made it clear in the final version of our guideline - which we issued in non-statutory form - that syndicated loans are not necessarily riskier than other types of lending. Indeed, the distribution of lending among participants is designed to reduce risk by sharing it. However, the very fact of sharing and the multiple parties involved can give rise to risks of their own - particularly of a legal nature - which means that the documentation in syndicated loans needs to be as watertight as possible.

The benefits of standardised documentation

This is why the Association's initiative in producing a suite of standard loan documents is so important, and is probably the Association's major achievement to date. Good documentation is not glamorous and is generally not fascinating reading, but it plays a vital role in underpinning the growth and liquidity of markets in financial products.

This is not a theoretical proposition. The explosive growth in the derivatives market is to a large extent due to the role played by the International Swaps & Derivatives Association (ISDA) in the development and maintenance of its derivatives documentation. The most recent testimony to this is the emergence within the space of the last few years of a US$1 trillion credit derivatives market, built on the platform provided by the Standardised ISDA Default Swap Documentation.

Why is standardised documentation such a powerful driver of market growth? The answer is that it creates more predictability and certainty about the characteristics of the financial contract in question. People are better able to understand what their rights and obligations are under the contract, and this gives them more confidence to enter the market and transact.

I would therefore encourage as many market participants as possible to begin to use the Association's new documentation. This should help to boost the liquidity of the primary market in syndicated loans, but it should also serve to increase turnover in the secondary market which is still relatively underdeveloped in Hong Kong. It is easier to trade a product whose terms and conditions are homogeneous.

The Association's intention to move on in the second phase of its project to develop standardised documentation for secondary loan trading will give further impetus to this market. This is something that the HKMA would particularly welcome since it would give banks greater flexibility to manage their loan portfolios and adjust their liquidity.

Apart from its impact on the market as a whole, standardised documentation has clear benefits at the level of the individual firm. Lending banks are better protected against the risk that contracts will turn out to be unenforceable, and that they will be left with a credit exposure that they did not anticipate. Moreover, the whole process of negotiating loans becomes more streamlined. There should not be a need to haggle over the standard "boilerplate" of the loan agreement, so that resources can be freed up to concentrate on the issues that are unique to the particular transaction.

The reduction in the number of idiosyncratic terms and conditions also means that the loan should be easier and safer to administer after the loan agreement is signed. Operational risk is therefore reduced.

Nor are the benefits one-sided. Borrowers will benefit from documentation that is easier to understand because it is written in (relatively) plain English and strikes an appropriate balance between the interests of lender and borrower. I understand that the Association's documentation embodies both these concepts. It is important that the borrower community should buy into the new documentation. In this connection, it is reassuring to know that the new documents are based substantially on those of the Loan Markets Association (LMA) in London. These have already been widely adopted in the European loan market and were developed in association with the UK Association of Corporate Treasurers. I believe that it is also intended to seek input from the borrower community in Hong Kong to help further develop the documentation.

Standardised documentation can therefore be described almost unreservedly as a good thing. This is subject to only two provisos. The first is that the basic documents will still need to be tailored to suit individual transactions - and the documents leave gaps for this to be done. This means that banks should not be lulled into a false sense of security. Legal advice will still be required for individual transactions, which should be good news for at least some of the people in this room. Second, the standard terms will need to evolve and be revised to take account of changing market circumstances and particular issues that arise. They may on occasions need to be tested in the courts. The recent experience of the credit derivatives market shows that ambiguities can arise even in standardised documentation, and the sponsors and users of the documentation need to be alert to this and make changes where necessary.

The link with operational risk

But the bottom line is that standardisation will reduce the risk that something goes wrong with the documentation. This feeds in turn into reduced legal risk, which is included within the definition of operational risk adopted by the Basel Committee. This defines operational risk as "the risk of loss resulting from inadequate or failed internal processes, people or systems or from external events".

This definition focuses on the causes of operational risk - internal processes, people, systems or external events. Operational risk losses can be further classified into various "event types" and assigned to each of a bank's major business lines. By classifying losses in this way, banks are better placed to identify appropriate risk controls and to assess the impact of these. In its draft paper on Sound Practices for the Management and Supervision of Operational Risk, the Basel Committee has identified seven main loss event types for operational risk. These include losses related to "execution, delivery and process management" within which resides documentation risk.

Why has the Basel Committee decided to focus on operational risk? The answer lies in the growing diversity and complexity of banking operations arising from well-known phenomena such as globalisation, the growing sophistication of financial technology and outsourcing. This has shown up in concrete form in the severe operational losses suffered by a number of international institutions. The Committee has therefore decided to take the lead in promoting greater understanding of operational risk and more consistency in the way in which it is measured, managed and controlled.

This would by itself be relatively unobjectionable. However, the Committee has decided to put some teeth into its recommendations by proposing an explicit capital charge for operational risk in the New Capital Accord. This is controversial for two reasons. First, there is no firm consensus on how to measure the capital required to support operational risk - the Committee is currently proposing three approaches to this, which range from the very basic to the quite sophisticated. Second, the amount of capital charge that the Committee originally had in mind was considered to be much too high. In response to this, the Committee is proposing a reduction in the proposed target of operational risk capital from 20% to 12% of minimum capital. A further reduction would potentially be available to those banks which use the more advanced measurement approaches.

This is an example of how capital requirements can be used as an incentive to banks to improve their risk management practices. This, in our view, is where the real value of the New Accord lies. Among other things, the increased focus in the New Accord on internal ratings and risk-adjusted pricing might encourage a more sensible approach to the setting of margins on syndicated loans. I know that many banks in Hong Kong think that margins are now too low to fully compensate for expected loss, the cost of capital and operational costs. However, more rational pricing policies are currently hindered by the lack of consistency among banks on how they price risk and allocate capital. The New Accord should go some way to correct this. Under the internal ratings based (IRB) approach, banks will be required to use internal ratings as a factor in the pricing of credit risk.

Progress on the New Accord

As part of the process of creating incentives for better risk management, the Basel Committee is now engaged in a more general effort to recalibrate the New Accord to adjust the amount of capital charge under each of the main approaches. The aim of the revised set of proposals is that, on average, the total amount of capital required under the standardised approach to cover both credit and operational risk should be the same as that required under the current Accord to cover credit risk only. Modest capital relief would be available to those banks that move to the foundation IRB approach, with still more relief for those which opt for the advanced IRB approach.

These, and other revised proposals of the Basel Committee, seem to have got the New Accord back on track. As you know, the project has suffered significant delays, and only a few months ago the chances of it going ahead seemed to be only 50-50. However, a number of the logjams have now been cleared and a revised timetable has been announced. The intention is now to release a revised set of consultation papers in May of next year. This would take account of a further quantitative impact study which will be held later this year to test the calibration of the revised proposals. If all goes well, the final Accord would be released before the end of 2003 with implementation targeted for 2006.

I know that there have been a number of false alarms on the New Accord, and that some of the banks here today may be inclined to take a wait-and-see attitude. To such banks, I would say that the pendulum has swung in favour of implementation. It would therefore be sensible to plan on the basis that the New Accord will now go ahead on the revised schedule, and that it will include an explicit capital charge for operational risk.

This does not leave all that much time for those banks which might want to use the advanced approaches in the New Accord to measure credit or operational risk to get themselves ready. I would advise such banks to start their preparations now if they have not already done so.

I would particularly recommend that banks begin to collect default data for credit risk and loss data for operational risk. A framework for doing the latter is set out in the Basel Committee's Sound Practices paper in the form of the two-dimensional matrix I described earlier, which assigns losses arising from the various loss event types to the major business lines of the bank.

Conclusions

Operational risk has many facets, and the events which give rise to it cover a huge range from high-frequency low-impact at one end to low-frequency high-impact at the other. There is a correspondingly wide range of controls and techniques that can be used to prevent or at least mitigate losses. Sound documentation lies at the more bread and butter end of the spectrum of such controls and techniques. But it is no less vital for all that. The Association is therefore to be congratulated for its initiative in putting in place an essential building block in the form of the standardised documentation that is being launched today. This will help to promote the development of the syndicated loan market in Hong Kong and in the region more generally, and will also serve the objective of enabling banks to better manage their operational risk.

Hong Kong Monetary Authority
24 May 2002