ent creditworthiness of the borrower, its external credit rating, the collateral offered,
or the covenants extended.^4 Since the capital requirement was set too low for high-
risk/low-quality business loans and too high for low-risk/high-qualtiy loans, the mis-
pricing of commercial lending risk created an incentive for banks to shift portfolios
toward those loans that were more underpriced from a regulatory risk capital per-
spective; for example, banks tended to retain the most credit risky tranches of secu-
ritized loan portfolios.^5 Thus, the 1988 Basel Capital Accord had the unintended con-
sequence of encouraging a long-term deterioration in the overall credit quality of
bank portfolios.^6 The proposed goal of the new Basel Capital Accord of 2002 (BIS
II)—to be fully introduced, if approved as proposed, in 2006—is to correct the mis-
pricing inherent in BIS I and incorporate more risk sensitive credit exposure meas-
ures into bank capital requirements.^7
Hammes and Shapiro (2001)^8 delineate several key drivers motivating BIS II:
- Structural changes in the credit market.Regulatory capital must reflect the in-
creased competitiveness of credit markets, particularly in the high-default-risk
categories; the trading of credit risk through credit derivatives or collateralized
loan obligations; modern credit risk measurement technology; and increased liq-
uidity in the new credit risk markets. - Opportunities to remove inefficiencies in the lending market.In contrast to the
insurance industry, which uses derivatives markets and reinsurance companies
to transfer risk, the banking industry is dominated by the “originate and hold”
approach in which the bank fully absorbs credit risk. - Ballooning debt levels during the economic upturn, with a potential debt serv-
icing crisis in an economic downturn.For example, in 1999, debt-to-equity ra-
tios at Standard & Poor’s (S&P) 500 companies rose to 115.8% (as compared to
84.4% in 1990) and to 95% (as compared to 72% in 1985) ratio of household
debt to personal disposable income.^9
BIS II follows a three-step (potentially evolutionary) paradigm. Banks can choose
among (or, for less sophisticated banks, are expected to evolve from) the basic (1)
Standardized Model to the (2) Internal Ratings–Based (IRB) Model Foundation Ap-
proach to the (3) Advanced Internal Ratings–Based Model. The Standardized Ap-
proach is based on external credit ratings assigned by independent ratings agencies
3 • 2 BIS BASEL INTERNATIONAL BANK CAPITAL ACCORDS(^4) An indication of BIS I’s mispricing of credit risk for commercial loans is obtained from Flood (2001)
who examines the actual loan loss experiences for U.S. banks and thrifts from 1984–1999. He finds that
in 1984 (1996) 10% (almost 3%) of the institutions had loan losses that exceeded the 8% Basel capital
requirement. Moreover, Falkenheim and Powell (2001) find that the BIS I capital requirements for Ar-
gentine banks were set too low to protect against the banks’ credit risk exposures. See ISDA (1998) for
an early discussion of the need to reform BIS I.
(^5) For a discussion of these regulatory capital arbitrage activities, see Jones (2000).
(^6) However, Jones (2000) and Mingo (2000) argue that regulatory arbitrage may not have been all bad
because it set the forces of innovation into motion that will ultimately correct the mispricing errors in-
herent in the regulations.
(^7) The original timeline has been pushed back. The final draft of the proposals is scheduled for 2003,
with possible implementation in 2006.
(^8) p. 102.
(^9) The Federal Housing Authority reported that the percentage of homeowners whose mortgage pay-
ments were more than 30 days late exceeded 10% for the first time ever as of the first quarter of 2001
(Leonhardt, 2001).