Abstract

 


 



A Non-Cyclical Capital Adequacy Rule and the Aversion of Systemic Risk


Raphael Douady


Riskdata; CES Univ. Paris 1

December 7, 2009


Abstract:     
We present in this note a method for computing the regulatory capital of financial institutions, along with the Basel Committee requirements, which avoids the pitfalls of the Value-at-Risk and, in particular, the fact that – as observed during 2008 crisis – it aggravates systemic risk rather than preventing it. The computation is based on stress testing, with the following principles: (i) market scenarios are defined by the regulator; (ii) sensitivities are estimated by each institution, as well as the impact of scenarios defined by the regulator and reported to it; (iii) the regulator not only counts the number of violations of the risk reporting but also their size; (iv) the regulatory capital is a multiple of the worst stress test, where the multiplier depends on the size and the frequency of the violations. By letting the institutions estimate their sensitivities to extreme market shifts, the regulator not only avoids a costly burden, but also keeps institutions responsible for their reporting. On the other hand, by keeping control on the list of stress tests involved in the computation of the capital, the regulator offers itself a very strong lever to prevent speculative bubbles, by making them costly in terms of capital requirements.

Number of Pages in PDF File: 8

Keywords: Capital Adequacy, Value-at-Risk, Risk based Capital, Basel II Accords, Compliance

JEL Classification: G18, G28, E63

working papers series


Download This Paper

Date posted: May 29, 2010  

Suggested Citation

Douady, Raphael, A Non-Cyclical Capital Adequacy Rule and the Aversion of Systemic Risk (December 7, 2009). Available at SSRN: http://ssrn.com/abstract=1617419 or http://dx.doi.org/10.2139/ssrn.1617419

Contact Information

Raphael Douady (Contact Author)
Riskdata ( email )
6, rue de l'Amiral Coligny
Paris, 75001
France
HOME PAGE: http://www.riskdata.com
CES Univ. Paris 1 ( email )
106 bv de l'Hôpital
Paris, 75013
France
Feedback to SSRN (Beta)


Paper statistics
Abstract Views: 506
Downloads: 102
Download Rank: 133,200

© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright
This page was processed by apollo2 in 0.328 seconds