Reuters reports:
The U.S. Securities and Exchange Commission is ending its program to supervise large independent investment banks now that the five participants have collapsed or reorganized.
… while Dealbreaker handles the stage directions:
SEC officials mount their horses, tip their hats, and ride off into the sunset. Pan back to show village burned to the ground and citizenry slaughtered, voiceover by Wilfred Brimley waxing poetic, “They did what they came to do. Their work here was done.”
The official SEC Press Release states:
The last six months have made it abundantly clear that voluntary regulation does not work. When Congress passed the Gramm-Leach-Bliley Act, it created a significant regulatory gap by failing to give to the SEC or any agency the authority to regulate large investment bank holding companies, like Goldman Sachs, Morgan Stanley, Merrill Lynch, Lehman Brothers, and Bear Stearns.
So let the finger-pointing begin! The SEC Inspector General has released two reports on the matter; the first, titled SEC’s Oversight of Bear Stearns and Related Entitites: Broker-Dealer Risk Assessment Program is a classic of its genre – there wasn’t enough box-ticking, so everything went wrong. Accordingly, the Inspector-General has recommended additional box-ticking.
There is more meat in the second report, titled SEC’s Oversight of Bear Steams and Related Entities: The CSE Program, which, interestingly, has been liberally sprinkled with redactions.
There is a bias in the report which must be borne in mind when formulating policy, stated on page 9 of the PDF as:
Bear Stearns was com liant with the CSE program’s capital and liquidity requirements; however, its collapse raises questions about the adequacy of these requirements;
While I agree that such questions have been raised, they are irrelevant and should be consigned to the trash bin. It should not be the purpose of regulation to ensure that nothing will ever collapse. The proper purpose of regulation in the capital markets should be to ensure that collateral damage in the event of such a collapse is minimized and does not lead to systemic failure.
I will certainly agree that there is evidence that the BSC bankruptcy managed to achieve the potential for collateral damage and contagion, but when examining the apparent failure of regulation to prevent this occurance, it must be borne firmly in mind that regulators should not care a whit whether the firm goes bust, subordinated debt-holders lose money and everybody loses their jobs. Such events are part of business and an attempt to regulate them out of the realm of possibility will ultimately hurt the economy more than it helps.
If, however, it can be shown (or at least persuasively argued … I don’t want to set the bar too high!) that the Treasury guarantee of the assets was absolutely required in order to save the system, THEN we have a failure of regulation which should be examined for potential improvements.
Bear Stearns’ increasing reliance on secured funding indicates that, although it appeared to be compliant with CSE program’s capital requirement, the market did not perceive it to be sufficiently capitalized to justify extensive unsecured lending. In this sense, Bear Stearns was not adequately capitalized.
These facts illustrate that although Bear Stearns was compliant with the CSE program’s ten percent Basel capital requirement, it was not sufficiently capitalized to attract the funding it needed to support its business model. Although the Commission has maintained that liquidity (not capital) problems caused Bear Stearns’ collapse, this audit found that it is entirely possible that Bear Stearns’ capital levels could have contributed to its collapse by making lenders unwilling to provide Bear Stearns the funding it needed.
The fact that Bear Stearns collapsed while it was compliant with the CSE program’s capital requirements raises serious questions about the adequacy of the CSE program’s capital ratio requirements.
Well, no it doesn’t, as I asserted above. The fact that Bear’s collapse due to liquidity issues while it was compliant with capital requirements HAD SYSTEMIC IMPLICATIONS is what raises serious questions about the adequacy of the CSE programme’s capital ratio requirements.
To summarize, as early as November 2006, Bear Steams was implementing a more realistic approach to liquidity planning than contemplated by the CSE programsy liquidity stress test. While this more realistic approach may have helped Bear Steams in the summer of 2007, it was not sufficient to save the firm in March 2008. Bear Steams’ initiative to line up secured funding indicates that the crisis which occurred in March 2008 was not totally unanticipated by Bear Steams, in that Bear Steams had been taking specific steps to avoid such a crisis for more than a year before it occurred.
According to the expert retained by OIG in conjunction with this audit, the need for Basel IIfirms to undertake specific efforts to line up committed secured funding in advance of a stressed environment depends on the extent to which the Basel I1firms can rely on secured lending facilities from the central bank during a liquidity crisis. On the one hand, if it is assumed that secured lending facilities will always be available from the central bank, lining up committed secured lending facilities is not necessary. In this case, a liquidity stress test, which assumes that secured lending facilities will automatically be available is appropriate. On the other hand, if it is assumed that collateralized central bank lending facilities might not be available during a time of market stress, Basel II firms have incentives to line up committed secured lending facilities, in advance, from other sources. In the context of CSE firms which are not banks, the policies of the Federal Reserve towards making collateralized loans to non-banks becomes an important element of their liquidity planning process.
In the heavily redacted section detailing Finding 2; that [the SEC] did not adequately address several significant risks that impact the overall effectiveness of the CSE programme; the report states:
Bear Stearns had a high concentration of mortgage securities. Prior to Bear Stearns becoming a CSE, TM was aware that its concentration of mortgage securities had been steadily increasing.
…
Yet, notwithstanding [redacted] and warnings in the Basel standards, TM did not make any efforts to limit Bear Steams’ mortgage securities concentration.
Further, a leverage limit is recommended for the future:
Although banking regulators have established a leverage ratio limit, the CSE program has not established a leverage ratio limit. The adoption of leverage limits must be reassessed in light of the circumstances surrounding the Bear Steams’ collapse, especially since some individuals believe that this policy failure directly contributed to the current financial crisis.
I note with amusement that in this official review of risk management and supervision thereof, Wikipedia is cited as a source for a definition. Really! Page 20, note 110. Get with the programme, guys – Wikipedia is not an authoritative source.
Model validation personnel, modelers, and traders all sat together at the same desk.”‘ According to the OIG expert, sitting together at the same desk has the potential advantage of facilitating communication among risk managers and traders but has the potential disadvantage of reducing the independence of the risk management function from the trader function, in both fact and appearance.
This is really bad, a violation of the most basic principles of internal risk control.
In 2006, the expertise of Bear Steams’ risk managers was focused on pricing exotic derivatives and validating derivatives models. At the same time, Bear Steams’ business was becoming increasingly concentrated in mortgage securities, an.area in which its model review still needed much work. The OIG expert concluded that, at this time, the risk managers at Bear Steams did not have the skill sets that best matched Bear Steams’ business model.
And that part’s just bizarre! The concept is, however, endemic in the industry … ‘Hey, Fred! You’re doing Preferred Shares this week!’
Furthermore, the OIG expert believes that meaningful implementation of high grade and high yield mortgage credit spread scenarios requires both a measure of sensitivity of mortgage values to yield spreads as well as a model of how fundamental mortgage credit risk factors make yield spreads fluctuate. These fundamental factors include housing price appreciation, consumer credit scores, patterns of delinquency rates, and potentially other data. These fundamental factors do not seem to have been incorporated into Bear Stearns’ models at the time Bear Stearns became a CSE.
… doesn’t look like they were much good at quant work …
When selling an asset, Tier 1capital is reduced by the amount of losses on the sale, but capital requirements are also reduced by removing the asset from the bank’s portfolio. A bank looking to improve its Basel capital ratios by selling assets therefore has a perverse incentive not to sell assets that have modest capital requirements relative to the markdowns the banks should have taken but has not yet taken. This perverse incentive tends to amplify the tendency for markets to freeze up and become illiquid by reducing trading volume that would othennrise occur as banks sell losing positions into the market. On the one hand, these perverse incentives are mitigated to the extent that capital requirements on such assets are high and valuations are appropriately conservative. For assets that face a 100% capital haircut, for example, the bank gains no improvement in its capital ratios by avoiding taking a markdown, and the bank increases its capital by the proceeds of any asset sales. On the other hand, these perverse incentives are worsened to the extent that supervisors allow banks to avoid marking assets down quickly enough, to avoid taking appropriate valuation adjustments in a timely manner, or to understate assets’ risks.
Similar to what Dealbreaker claimed yesterday.
There is much of interest in the report; but a lot of the juicy stuff has been redacted, presumably because it was provided to the SEC on a confidential basis.
Update, 2008-10-7: Via Dealbreaker and Bloomberg comes some juicy stuff about all the redactions:
An unedited version of the 137-page study posted to Grassley’s Web site Sept. 26 showed that Bear Stearns traders used pricing models for mortgage securities that “rarely mentioned” default risk. A link on the site to the full report wasn’t working today.
The firm lost one a top modeler “precisely when the subprime crisis was beginning to hit” and writedowns were being taken, the full report said. “As a result, mortgage modeling by risk managers floundered for many months,” according to the unedited document, quoting internal SEC memos from April and December 2007. The comments were removed from the edited version publicly released by the SEC.
…
Trading and Markets unit members saw that Bear Stearns traders dominated less-experienced risk managers, the inspector general reported in sections that were excised from the public report.
“As trading performance remained strong for years in a row, it clearly wasn’t career-enhancing to stand in the way of increasingly powerful trading units demanding more balance sheet and touting their state of the art risk-management models,” said Brad Hintz, an analyst at Sanford C. Bernstein & Co. in New York, and a former chief financial officer at Lehman Brothers Holdings Inc.
In other words, the risk managers were there as part of the standard box-ticking exercise, not because anybody in management really wanted them to do anything. I noted on April 16 one particularly nasty report with respect to a Merrill Lynch CDO offering of a corporate finance guy bullying a trader to make his underwriting go.