Prudential Supervision: What Works and What Doesn't

Abstract

This excellent book is the edited proceedings of an NBER conference held in the USA in January 2000, (all those involved should be congratulated on the speed of publication), with the addition of an introduction by the editor, Mishkin. This was an extremely well designed conference, tightly focused, with a small and carefully chosen set of participants, being academic and practitioner experts in banking regulation. There is a general suspicion that academics will reserve their best work for journals, leaving only slightly sub‐standard papers for conferences, and their proceedings, but not so in this case; in my view all the seven papers reprinted here are of high journal standard class. Moreover they each have what journals lack, which is the addition of a discussant's commentary, and these again were uniformly excellent, perceptive and informative. The seven papers are as follows: an econometric cross‐country survey, from the World Bank team of Barth, Caprio and Levine, of the relationship between restrictions on bank activities (i.e. securities, insurance and real estate) and on ownership and crisis, with a finding that more restrictions are associated with more crises; and also of the relationship between the prevalence of publicly owned banks and crisis, again a positive relationship. The second paper, by Bliss and Flannery, is an econometric study aiming to examine whether bank managers appear to react to abnormal falls in their debt or equity prices. While the result is negative, a general concern is whether one can set up such a test unambiguously. The third paper, by Calomiris and Powell, is an historical description of the reformed regulatory framework for banks in Argentina, (1992–9). This is given greater weight by the close involvement of its authors with the introduction and operation of that framework, Powell then being chief economist at the Central Bank of Argentina.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 134 citations exist upstream.