Hollywood loves to make sequels to successful movies and develop them into franchises. Historically, this has been met with varying degrees of success (think of Batman and Robin versus The Dark Knight Rises). Now, it seems as if prudential bank regulators might be taking a lesson from Hollywood’s playbook.

The ink was barely dry on the Basel II accord when the financial crisis hit in 2008. In its aftermath, the global banking rule-maker, the Basel Committee on Banking Supervision (BCBS), scrambled to raise not only the quantity of capital banks had to hold but also its quality—less complicated hybrid debt and equity instruments and simpler common equity types (the latter with higher loss-absorbing characteristics and more flexibility to pay dividends). Now regulators are investigating what appears to be the next evolution in bank capital standards: consistency in the calculation of risk-weighted assets.

Where it all began

The first iteration of the Basel rules was drawn up in 1988 and implemented in 1992. It introduced the concept of “risk-weighting”—reducing the value of an asset for capital calculation purposes to reflect the relative riskiness between different assets. Risk weights within specific classes of assets were then standardized, thus dictated by the regulator. For example, government debt of developed countries would receive a risk weighting of zero percent, and no capital would be held against it. In contrast, an unsecured loan to a consumer would have a 100 percent risk weight, and the bank would therefore need to hold 8 percent capital against the full amount outstanding. Initially, the Basel I rules only accounted for credit risk. Then, in 1997, Basel published additional requirements to hold capital for market risk.

For a decade, these rules worked well. However, the business of banking became much more globalized and complex. It became apparent that standardized risk weights were a blunt tool to measure a bank’s risk, given the diversity in the types of assets a bank may hold in its portfolio.

After several years of study and negotiation, the BCBS published Basel II, which refined the methodology that banks can use to calculate risk weights by allowing them to estimate their own. Under these rules, banks use their internal data for credit and operational losses, and they use value at risk for market risk to predict future losses. Banks with lower historical losses would be rewarded with lower risk weights and therefore required to hold a lower amount of capital. However, the 2008 financial crisis proved that many banks around the world did not hold enough capital, and governments had to provide extraordinary funding to prevent many of them from failing.

Basel III was the BCBS’s response to the crisis, introducing new standards for capital, liquidity, and leverage. It required banks to significantly increase the amount of capital held as a percentage of risk-weighted assets and to improve quality by increasing the required proportion of capital such as common shares and retained earnings. Basel III also introduced capital buffers to be built up in times of credit expansion and drawn down in a bust, as well as surcharges for banks that have global or national systemic importance. While no banks in Canada were identified as being both globally and systemically important, the big six Canadian banks, the Caisse de dépôt et placement du Québec, and Central 1 Credit Union were identified as being systemically important to Canada.

While Basel III focused on improving capital, it remained largely silent on risk-weighted assets. In January 2013, the BCBS published a study that sampled global banks’ methodology for calculating risk-weights. The study estimated the ratio of risk-weighted assets to total assets and found wide ranges between banks. Not surprisingly, this was explained primarily by differences in the underlying assets themselves. In order to control for this, the BCBS also provided a hypothetical test portfolio of assets to a sample of 15 banks (one of which was Canadian) and asked them to compute risk weights using their models. This hypothetical portfolio exercise found a substantial difference between the bank reporting the lower risk-weighted assets and the bank reporting the highest. It attributed this both to the rules that applied to their respective national jurisdiction and also to the modelling assumptions made by the bank. This last point is of particular importance to the BCBS.

This isn’t an unexpected outcome when one considers the inputs into certain bank exposures.

For example, how would one calculate the probability of default, which is based on historical loss information, for an exposure to a particular country that has never defaulted on its debt? Assumptions are therefore made to fill in the blanks. Bankers will attempt to make an educated guess about this, but biases will always creep in.

Since their first study, the BCBS has published two more reports on the issue, concluding that there is in fact inconsistency in banks’ risk weights methodology. Several other organizations, including the European Banking Authority and KPMG, have also highlighted this issue. The Financial Stability Board published a report stating, “The risk models that banks use to calculate their capital show worryingly large differences.” Indeed, the chair of the BCBS has highlighted this issue in several speeches, indicating that the BCBS is open to the use of benchmarks and floors for certain parameters—back to standardized values once again.

However, not every regulator is convinced. In an October 2013 speech to the Risk USA Conference in New York City, Mark Zelmer, Deputy Superintendent of Canada’s Office of the Superintendent of Financial Institutions, indicated that returning to a less risk-sensitive measure by introducing standardized values will result in greater demands on their bank examiners. He argued, “Simpler capital rules would require even more intensive supervision of individual banks.” It would appear that gaining consensus in the regulatory world might be more difficult than first thought.

The stage is being set for another round of changes to the Basel framework. Let’s hope this next instalment in the franchise isn’t just another reason to make another movie.