Third in a nine-part series

The Basel Committee recently published for the first time an analysis of risk-weighted assets of banks' credit risk. In doing so, the committee confirmed the worst kept secret in the industry: there is significant variability in banks' RWAs especially between large U.S. and European financial firms.

The findings demonstrate that most of this variability is due to differences in riskiness of assets. In this environment, however, many pundits are skeptical of the well-meaning Basel committee's claim that variances in RWAs were not due to banks' manipulation. Some have even accused the committee of being too timid in policy prescription.

I certainly believe that banks can and do manipulate inputs. Yet, I have also seen significant variances in RWAs because models are not a science; they are an art. Their results are influenced by the availability, amount and quality of data, macro-assumptions used and, often, the extent of management intervention in how models are constructed and used.

Also, it is important to recognize that variability in banks' RWA calculation is not only due to bank's modeling, but also due to supervisory choices in each jurisdiction, driven either by discretion permitted under the Basel framework or deviation in how a country's regulators implement the Basel accords.

Certainly, globally systemically important banks are not helping themselves by continuing to be opaque about what assumptions they use to derive their models' credit risk drivers. It is understandable that significant distrust about banks will continue, particularly when most people have never seen or used a model. I have discussed previously in these columns that Pillar III, the part of Basel that provides guidelines for disclosures, is a very important component of the Basel accords as it is the only pillar within the Basel accords useful for market discipline.

Pillar III has existed since 2004. Yet, nine years later, U.S. banks are still not compliant. In numerous countries such as Brazil, Canada, India, most of continental Europe and the U.K., GSIBs do publish Pillar III disclosures, but those disclosures are neither robust nor uniform enough, making peer comparisons difficult.

Unfortunately, due to the Basel Committee's focus on Pillar I and II, updating and strengthening Pillar III continues to be neglected. The July 5th consultative document highlights that the Basel Committee's medium-term goals are potentially to harmonize RWA implementation requirements and place constraints on parameters used by advanced internal ratings-based banks. Unfortunately, the committee did not make it clear as to when potential harmonizing guidelines would be forthcoming.

Banks should be required to give more disclosure on the types of assets that they have on- and off-balance sheets, especially because off-balance sheet instruments usually end up getting lighter capital treatment. It would also be useful to know what internal ratings distribution and associated risk parameter estimates a bank has and its share of defaulted exposures. Presently, it is practically impossible for bank observers to see what percentage of a bank's internal ratings represent high or low credit quality and what each rating represents in terms of default probability.  For investors and bank analysts, information about the major sources of changes in RWAs over reporting periods and credit risk approaches would be useful to understand the construction and reliability of RWAs and to be able to compare banks.

Also impacting RWAs is how derivatives get translated to on-balance-sheet-item equivalents by a system of credit conversion factors. Identifying banks' derivatives positions and their role in RWAs should not require a degree in forensic accounting. To understand RWAs, it is imperative for banks to disclose if they are on the buying or selling side of a derivatives transaction. As a protection seller in an option or credit default swap, an institution can become illiquid almost instantly when market forces rapidly move against it. Given how material this contingency liability is for potential and existing investors, this information should be made publicly available in a much more digestible form than the Federal Reserve's FR Y-9C Schedule L.

Importantly, banks should disclose that the netting of derivatives positions, as shown by Office of the Comptroller of the Currency data, allows them to show a much lower credit exposure, which leads to lower regulatory capital. If at any point netting contracts were suddenly disallowed or failed to function properly in a time of stress, a bank's credit exposure could increase at barely a moment's notice and banks would end up not being sufficiently capitalized to sustain unexpected losses.

Also, banks need to explain more where they house their derivatives and the associated collateral. These characteristics are extremely important to how RWAs can change and also in the event that regulators must liquidate a bank. Bankruptcies are dictated by where the subsidiaries are, not where a bank is headquartered.

There will always be market participants who will never like or trust the RWA framework. Yet, so long as RWAs are an integral part of capital calculation, domestic bank regulators must demand more disclosure about key credit risk drivers and a good description about on- and off-balance-sheets risks. It is also important for observers and analysts to see how a banks' credit exposure due to derivatives can change.

For their part, investors and journalists have to do more than just say that RWAs "are complex" or "can be gamed." They need to learn about the RWA framework enough to demand relevant information from banks. Investors who do not like the information they receive can vote with their feet. In learning more about RWAs, journalists can make substantial contributions to the market by asking better-informed questions at earnings calls or conferences.

Using a capital charge reliant only on RWAs is definitely not enough to keep banks well capitalized. The Basel Committee certainly understands this as it has added key leverage, liquidity and SIFI buffers in the 2010 Basel Accords and provided updates for most of those buffers in the last few weeks. Yet, until banks' RWA disclosures improve and become more uniform globally, capital ratios will remain a mystery to many journalists, investors, banks and even some regulators.

Next: A look at capitalizing derivatives' counterparty risk exposures.

Mayra Rodríguez Valladares is managing principal at MRV Associates, a New York-based capital markets and financial regulatory consulting and training firm. She is also a faculty member at the New York Institute of Finance.