Page 327 - untitled

 FINANCIAL CRISIS INQUIRY COMMISSION REPORT

different scenarios that matched its risk profile, the supervisors tried to maintain
comparability between the tests. The tests assumed that each firm would lose  of
unsecured funding and a fraction of repo funding that would vary with the quality of
its collateral. The stress tests, under just one estimated scenario, concluded that
Goldman Sachs and Morgan Stanley were relatively sound. Merrill Lynch and
Lehman Brothers failed: the two banks came out  billion and  billion short of
cash, respectively; each had only  of the liquidity it would need under the stress
scenario. 
The Fed’s internal report on the stress tests criticized Merrill’s “significant amount
of illiquid fixed income assets” and noted that “Merrill’s liquidity pool is low, a fact
[the company] does not acknowledge.” As for Lehman Brothers, the Fed concluded
that “Lehman’s weak liquidity position is driven by its relatively large exposure to
overnight [commercial paper], combined with significant overnight secured [repo]

funding of less liquid assets.” These “less liquid assets” included mortgage-related
securities—now devalued. Meanwhile, Lehman ran stress tests of its own and passed
with billions in “excess cash.” 
Although the SEC and the Fed worked together on the liquidity stress tests, with
equal access to the data, each agency has said that for months during the crisis, the
other did not share its analyses and conclusions. For example, following Lehman’s
failure in September, the Fed told the bankruptcy examiner that the SEC had de-
clined to share two horizontal (cross-firm) reviews of the banks’ liquidity positions
and exposures to commercial real estate. The SEC’s response was that the documents
were in “draft” form and had not been reviewed or finalized. Adding to the tension,
the Fed’s on-site personnel believed that the SEC on-site personnel did not have the

background or expertise to adequately evaluate the data. This lack of communica-
tion was remedied only by a formal memorandum of understanding (MOU) to gov-
ern information sharing. According to former SEC Chairman Christopher Cox,
“One reason the MOU was needed was that the Fed was reluctant to share supervi-
sory information with the SEC, out of concern that the investment banks would not
be forthcoming with information if they thought they would be referred to the SEC

for enforcement.” The MOU was not executed until July , more than three
months after the collapse of Bear Stearns.

DERIVATIVES: “EARLY STAGES OF ASSESSING
THE POTENTIAL SYSTEMIC RISK”
The Fed’s Parkinson advised colleagues in an internal August  email that the sys-
temic risks of the repo and derivatives markets demanded attention: “We have given
considerable thought to what might be done to avoid a fire sale of tri-party repo col-
lateral. (That said, the options under existing authority are not very attractive—lots
of risk to Fed/taxpayer, lots of moral hazard.) We still are at the early stages of assess-
ing the potential systemic risk from close-out of OTC derivatives transactions by an
investment bank’s counterparties and identifying potential mitigants.” 
The repo market was huge, but as discussed in earlier chapters, it was dwarfed by