|
Which banks are good at telling their shareholders and counterparties about risks they are running and how they manage them? The answer is, none. Not one of the world’s top global players discloses a satisfactory amount. Most feed pap to the readers of their annual reports about the nature of swaps and standard deviations without showing how they apply risk measurement to their own exposures, let alone saying what the numbers are. Geneva-based think-tank Ifci (International Finance & Commodities Institute) commissioned a study on bank disclosure based on their annual reports. Arthur Andersen volunteered to sift through the reports of 42 top international banks, including six top securities firms, and mark them according to an Ifci disclosure matrix (see below). The marking process was necessarily subjective, since it took into account language and explanations used. But the results send a strong, objective and depressing signal to the banks – they all fail to disclose enough, either because they haven’t grasped the risks themselves, or because they are masters of opacity. The results were so poor that Ifci and Arthur Andersen decided not to publish a detailed scoresheet in the hope that the banks will voluntarily do better next time in their 1999 reports. Naming and shaming might have sent a tougher message, but some of the criteria on which the banks were judged aren’t yet mandatory reporting requirements, or indeed relevant to banks not specialized in certain areas. Moreover, Arthur Andersen was reluctant to upset some of its major clients. However, the issue of better disclosure is high on the supervisors’ wish list and the pressure to disclose more will continue. The Basel Committee on Banking Supervision’s guidelines last June for a new supervisory framework made clear its view that more disclosure is an important way to impose better market discipline on banks. In October it published its recommendations on public disclosure, presenting a checklist of qualitative and quantitative categories it would like banks to explain and reveal. In January it published views on more transparency regarding the nature of a bank’s capital. It also did its own survey of 1998 annual reports which among other things noted huge variations in the amount of information given. Only a handful (about a quarter of the banks surveyed) attempt to quantify such risks as their liquidity risk, their potential future credit exposure, or the value-at-risk of their non-traded portfolios. Meanwhile the G7 has two initiatives on disclosure by financial institutions. One, led by the Federal Reserve Bank of New York, is asking big banks to disclose, non-publicly, their exposure and concentration risk with various sectors and counterparties. There are two objectives: first, to ensure that these institutions have a proper grasp of the risks they are running; second, to identify any concentrations of risk that could destabilize the financial system. Since banks are so reluctant to disclose any more information than they have to, asking them to disclose to regulators in confidence may be one route. But that puts the burden on regulators to spot and follow up weaknesses in banks and in the system. The beauty of a market-based disclosure discipline – and this is supported by the new Basel framework – is that shareholders and the market, rather than a bunch of bureaucrats, can force change and reward compliance. Ifci’s original plan, to name and shame, has only been put on hold. Ifci’s study was based on a matrix designed by Rajna Gibson, professor of finance at Lausanne University. It focused on market and credit risk as the easiest classes to measure across diverse financial institutions. Operational risk may be included in future surveys. No financial institution in the survey even managed a half-mark of 50%. Many institutions are still giving generic information about what an option or swap is. This does not constitute disclosure. Few firms distinguish between the risks on their trading and non-trading books. There are many cases where market risk is discussed purely in terms of the trading book and credit risk in terms of the banking book. North American banks generally have better disclosure practices than the Europeans – there are 16 North American banks in the sample and six make it into the top 10 for trading disclosure. That is mainly the effect of the more rigorous US GAAP. Non-trading disclosure is even poorer, but European banks tend to do a little better with five in the top seven – the two others being North American. The worst are the Dutch and the Germans. No bank gives a good general overview of its risk management policy, although some of the best sets of accounts give details of the roles and responsibilities of various individuals and committees. No bank mentions significant changes to its trading book and none gives a profit forecast. Nor is there an attempt to identify or discuss concentration risk across asset classes by geographical area or industry group. Virtually none of the banks mentions collateral and collateralization policy, and the related liquidity risks that should be taken into account. Collateral is generally lumped in with mention of netting agreements when net exposures are given. Trading positions and notional amounts have been given for derivatives in a lot of cases, but not for the full trading book. It could be argued that non-derivative transactions are immaterial to the trading book, but this should have been stated. Many banks seem to equate derivative positions with trading positions and this is a gross misrepresentation of the trading risks of some banks. Only four banks out of 42 mentioned that they had an internal credit risk model or were installing one. Virtually no back-testing information is given by any bank for non-trading book VaR. Some 26 out of the 42 banks scored 0 for this section There is clearly a big gap between the risk management that banks discuss between themselves, and what they disclose to the public. Either they don’t want to disclose it, or there is a communication problem. Says Gibson of Lausanne University: “We’ve seen the difference between what people claim, and what they’re doing.” Obviously there is a time lag between banks’ experience and producing their annual report, but it is the best common ground for comparison. Gibson hopes disclosure will be more timely in future: “Why can’t people do it quarterly via the internet?” David Shirreff Disclosure matrix Trading activities Overview of risk management policy Market risk VaR information Backtesting information Stress-testing information Credit risk Overview of credit risk in trading General description of trading position Current and potential credit risk exposures Internal credit risk modelling Integrating market and credit risk management Concentration risks Non-trading activities Overall disclosure of non-trading activities Market risk VaR for non trading activities Asset and liability management Hedging policy Credit Risk Overview of non-trading activities and credit risk policy Counterparty data analysis Concentration risks Cross-border analysis Source: Ifci |
Related
From share of wallet to share of trust
Sponsored by Santander