Multiple initiatives are due to be introduced this year to reform the international financial architecture in response to the global financial crisis that has erupted since mid-2007. Till now, statements emanating from governments, regulators and multilateral organisations detailing principles and action plans, though comprehensive, were mostly generic in nature. The current set of legislation wending its way through the US Congress, the Bank of International Settlement, the US Securities and Exchange Commission, the European Commission and other bodies will lay out detailed regulatory frameworks.

One of the key messages that seems to have emerged is the need for international coordination in the emerging global financial architecture. But what exactly would this entail? Based on our understanding of the risks that have emerged from the crisis, there are two issues that need greater thought and discussion. These seem to have been indirectly addressed in most official statements and documents like the communiqu? from the G-20 meet earlier in 2009 and the Report of the Financial Stability Forum on Enhancing Market and Institutional Resilience.

The first is the need for counter-cyclicality in computing regulatory capital for banks. One of the features of the Basel II system is the pro-cyclical aspect of its capital requirements for banks. The problem is as follows. During the downswing of a business cycle, as banks? asset quality deteriorates, banks need to raise more capital at a time when financial markets are less favourable to their being able to do so. Under the Internal Ratings Based approach of Basel II, capital requirements respond to changes in credit default risks [as measured by Probabilities of Default (PD) and Loss Given Default (LGD)]. Capital requirements thus will tend to increase in an economic downturn and fall during the upcycle. This risk sensitivity is enhanced by fair value accounting practices, introduced through International Accounting Standard 39 (IAS39), particularly mark-to-market (MTM) accounting. Although the utility of IAS39 for disclosure and market discipline is undeniable, MTM does reinforce pro-cyclicality. Instead, if capital requirements were raised during the upswing, banks would be better protected during the downturn, ensuring that banks? economic capital does not fall below the requirements of regulatory capital.

Various measures to introduce an element of counter-cyclicality need to be considered. The simplest means of addressing this is to get financial institutions to allocate more capital during upswings, based on the usually coterminous asset price inflation. One proposal is to use the regulatory discretion under Pillar 2 of Basel II to encourage banks to provide capital in a ?through-the-cycle? manner. This was done in 2006 and 2007 by RBI, which increased risk weights for assets in certain sectors and increased provisioning requirements for standard assets. These measures, however, are somewhat arbitrary. Greater insights into economic processes will be needed to establish capital rules ?through the cycle?, based on analytical models of growth and credit cycles.

The second is treatment of systemic risk and stress testing. The mix of Basel II, mark-to-market accounting and IAS39 has many desirable effects. They give each individual bank a much clearer and better-defined picture of its own individual risk position. Banks identify and mitigate risk by stress-testing for adverse market scenarios. The purpose of regulation should, however, also be to contain systemic risks, the possibility of contagion, and the externalities of the system as a whole. One of the lessons of the global financial turmoil is that however effective and robust the risk management systems of individual banks, they are likely to be completely subsumed by extreme systemic volatilities and risk. The systemic problem is that the action of each individual bank impinges on all other banks. It is then inadequate for banks to individually conduct stress tests. In hindsight, all assumptions of correlations and fat-tailed probability distributions that have hitherto been used for stress-testing and calibrating value-at-risk have been shown to be inadequate in estimating the extent of systemic risk. Systemic stress testing becomes more important.

Such systemic stress tests are best conducted by regulators and central banks. Coordination between central banks becomes important when systemic risk becomes globalised, being correlated across geographies. Systemic risk mitigation also requires a common view to emerge on desirable leverage for financial institutions and on the role of the (lightly regulated, or often unregulated) shadow banking system that grows through regulatory arbitrage. The problem thus requires international coordination. These generally are more in the domain of central banks and policy authorities of countries rather than individual banks and financial institutions. Uniformity across regulatory jurisdictions appears vital as otherwise more tightly capital-regulated banks would become uncompetitive for investors as their return-on-equity would fall.

The author is vice-president, business & economic research, Axis Bank. Views are personal