Comments

Patricia Correa: Endless pages have been written on the potential flaws that could prevent Basel II from achieving the ultimate goal of increasing financial stability worldwide by improving bank risk management and making capital requirements more sensitive to risk. Criticisms, which are sometimes contradictory, can be grouped into five categories. First, the new accord offers alternative approaches for measuring capital requirements (two versions of the standardized approach and two of the internal ratings-based, or IRB, approach), but it does not create proper incentives to use the most risk-sensitive approach, thereby opening the door for arbitrage. Second, in countries with little capital market and financial development, reliable external ratings are not available for most of the assets in the banks' credit portfolio; in such cases, the standardized approach will do little to link better capital to risk and would be, at best, a poor substitute to Basel I. Third, Basel II relies heavily on methodologies that lead to capital requirements that tend to accentuate the cycle, which can eventually increase, rather than diminish, financial instability. Whether this flaw can be attributed to the methodologies themselves or to the length of the databases employed to make the respective estimates is a controversial issue, but this topic is certainly an important one in the discussion. Fourth, the new framework will most probably increase the amount of minimum regulatory capital regardless of the measurement methodology chosen, and many banks, particularly in emerging economies, are simply not ready to meet the additional capital requirement. Finally, regulatory and supervisory bodies in most emerging economies are not prepared to meet the challenges posed under the second pillar of the accord, owing to lack of infrastructure, inadequate human capital, and so forth.

Majnoni and Powell's paper represents an important contribution to the understanding of the potential impact of Basel II on emerging economies and sheds light on the validity or relevance of the aforementioned criticisms. The authors' contribution is particularly valuable in two aspects: it is the first attempt, to my knowledge, to estimate capital requirements in Latin American economies using the same methodology, enabling comparison across countries; and the paper goes beyond simply criticizing the new accord and, within the spirit of Basel II, constructively proposes a new approach to capital requirements (namely, the centralized ratings-based approach). However, issues related to the estimation of recovery ratios, which represent a key component of capital requirement estimation, are left out of both the quantitative and qualitative analyses in the paper. This makes the analysis of the challenges ahead incomplete, and it leads to some erroneous conclusions about the relative advantages and disadvantages of Basel II over Basel I. The authors' recommendations are also somewhat unclear and self-contradictory regarding how far and how fast these countries should move toward implementing better risk management systems and more risk-sensitive capital requirements. Below I present more specific comments on each of the paper's three main components: a quantitative study estimating capital requirements that accomplish the objectives of Basel II in three emerging economies; a qualitative analysis of the practical difficulties of implementing Basel II, with an assessment of a country's readiness to adopt Basel II; and policy recommendations.

With regard to the empirical analysis, the authors use a bootstrap methodology—which they argue is free of the usual problems that plague traditional econometric techniques (namely, parameter estimation and model errors and risks)—to estimate default probabilities for the loan portfolio of Argentina, Brazil, and Mexico. They describe their specific assumptions about the provisioning system and their allegedly representative sample of credit data.¹ They then calculate the capital requirements that would be necessary to cover credit risk under two circumstances: first, using the foundation IRB approach and the default probability risk weight mapping curve proposed by Basel II; and second, using the bootstrapping methodology to calculate the capital requirements needed to cover the value at risk of those portfolios at a 99.9 percent confidence rate.

When comparing the two results, the authors conclude that...