A follow-up to "Is The CDS Market Manipulated?" by Joshua Rosner of Graham Fisher.
ISDA Determinations Committee’s "External Review", an Inside Job (pdf)
Last week we focused on potential manipulations of the opaque and self-regulated process by which the conflicted members of ISDA’s Determinations Committee (“DC”) determine whether a triggering event has occurred. This week we will focus on the inherent problems in the External Review process, as set out in the Determinations Committee’s rules.
While many would argue that self-regulation is the best approach to markets, this argument is baseless when 15 firms are able to, without independence and transparency, oversee an entire market and make enormously impactful decisions without disclosing conflicts or explaining their rationale to anyone. At the end of the day, buyers and sellers of protection who enter into CDS transactions after careful consideration of the contractual language could find themselves catastrophically exposed to significant financial losses as a result of a process that relies more on the whims of a conflicted few than on the rule of law.
On the surface, the External Review process appears clear, well defined and generally appropriate. In reality, the process reaffirms the power of the DC majority to control the outcome. Problems with the operation of the Determinations Committee are only reinforced in the External Review process. As a result, even in External Review, the determination of triggers on CDS remains in the hands of a few players who have awarded themselves regulatory license, are operating without strict prohibitions against conflicts of interest and appear to be encouraged to front-run and manipulate the market.
We do believe that an independent External Review process, if appropriately implemented and overseen, would render the inherently conflicted DC process superfluous. However, until problems with the External Review process are addressed, it should be viewed with the same reasoned suspicion as the DC process. In reviewing the External Review process rules, there are a number of potentially significant flaws that leave CDS holders exposed to further potential conflicts. These flaws prevent the External Review process from living up to its potential. Instead, market participants are left exposed to unforeseen risks in a world in which decisions are made by those who “have no duty to research, investigate, supplement or verify the accuracy of information on which a determination is based”, are “not obligated to follow previous determinations or to apply principles of interpretation such as those that might guide a court in interpreting contractual provisions”, who “may have an inherent conflict of interest in the outcome of any determinations”, and “may vote and take other actions without regard to your interests under a Credit Transaction”.
The Process:
Recently, we highlighted the myriad of problems with the process by which the Determinations Committees determine if CDS triggers have occurred. If one of the regional DCs is unable to come to a decision, by a supermajority of the 15 members, regarding whether an event has occurred, then they may refer it to an “External Review” process.
The External Review process begins with the formulation of the questions to be reviewed and decided on by this External Review panel. Once the question(s) have been agreed upon they are required to be “promptly published by the DC Secretary on its Website, along with the answer relating to each Presented Position and the Submission Deadline with respect to such Reviewable Question”. After the question(s) have been agreed upon and published on the website, the next step is the selection of External Reviewers.
The DC Secretary for each region is required to maintain a list of “Potential Pool Members”, who are nominated by the same firms that serve on the Determinations Committee, and to “publish the members of each External Review Panel List from time to time on its Website”. From the pool of available Reviewers, the DC votes for five – three will be panelists and two will act as alternates in case any of the other three are dismissed or resign from the panel.
Once the panel is selected, and within seven days of the referral of the Eligible Review Question to the DC Secretary, written materials supporting positions are to be submitted to the panel. At least two days after, but no longer than four days after written materials are submitted, the panel is required to hear oral arguments supporting the DC positions. Within five days of the deadline for submission of written materials, the panel is expected to render its decision.
The Problems:
Reviewable Question
In both External Reviews to date – CEMEX (2010) and Caesars (ongoing) – the reviewable question(s), which is required, by rule, to be promptly published on the website before a panel is selected and charged with a review, were not published “in a timely manner”.
Reviewer Selection
As we briefly noted last week, the notion that the External Reviewers are independent and unbiased is undermined by the manner in which they are selected as potential External Review pool members. While each DC member has the ability to nominate External Reviewers, the voting process to elect them to a panel is inherently flawed, biased and rife with potential conflicts of interest.
While the list of potential pool members is required, by rule, to be on the ISDA DC website, and updated from time to time, it is not and has never been. In fact, and as example, there was no public information about who served on the External Review panel for the CEMEX review until after the panel had reached a decisionv. This leads to questions about whether there were undisclosed conflicts among panelists and whether there was disagreement among the DC members regarding the nominated panelists.
When a question is to be decided by an External Review process, members vote on Reviewers and “may Resolve by unanimity to select up to five Potential External Reviewers”. This approach is clearly little more than a fig leaf. If, in fact, there were meaningful constraints on the ability of members to nominate parties that are biased toward them, then it would be expected that there would be few occasions in which there would not be unanimity. Instead, as we saw with CEMEX, there was clearly not unanimity. Where the DC cannot unanimously agree on five panelists then a simple majority can vote to allow the DC Secretary to choose the Reviewers needed to bring the panel to five. However, the majority can ensure that the entire pool from which the DC secretary chooses was selected by the majority, thus making a mockery of the idea that panelists selected by the DC secretary are truly objective or random. Given the inherent lopsided nature of the process, with dealers (more likely to be sellers of protection) holding 10 of the 15 DC spots, it seems likely that the DC Secretary would pick Reviewers amenable to the majority (which will almost invariably be led by dealers). Thus, the majority controls the process from start to finish.
In an ideal world these Reviewers, who are supposed to represent the interests all buyers and sellers of protection on CDS, should be selected in a public process, with rigorous and explicit policies intended to prevent conflicts of interest or bias and with a requirement that they publicly disclose their work histories, any income received by a member of the DC and existing holdings of any reference security or CDS. Unfortunately, no such requirements exist. The ISDA policy regarding conflicts of interest for External Reviewers does not require any such public disclosure of conflicts, nor does it require disclosure of any financial relationships between Reviewers and DC member firms. With no definition of terms, it only requires that Reviewers not be employees or affiliates of a DC member and potential Reviewers are required only to “disclose to the Convened DC any conflict of interest which exists or is foreseeable with respect to either the Reviewable Question or the related DC Questions which may be deliberated by the Convened DC”. Moreover, even where a DC member raises concerns about conflicts or bias, there does not seem to be any mechanism for those concerns to be addressed. Clearly, the External Review process is not independent and its protections against conflicts are threadbare. One merely has to look at past panelists and pool members to recognize this.
In the CEMEX External Review, there was no public disclosure, ahead of the panel’s decision, about who were potential pool members, potential panelists or those who were to be voted on. Only after the panel’s decision was rendered was it disclosed that there were only four panelists, not the five required. This suggests that either one panelist was dismissed or that there was a failure to unanimously select five panelists and the DC Secretary failed to choose the fifth. That failure to reach unanimous consensus might well have been the result of the obvious conflicts of interest among those who were selected. Unfortunately, the public will never know of the concerns or disagreements among DC members, in the selection of External Reviewers given that DC rules explicitly state, “the DC Secretary shall not publish the binding vote taken for purposes of confirming a Potential Pool Member”.
This should be of concern to regulators and the public given the example and conflicts within the CEMEX External Review panel. Of the four CEMEX panelists, one had been an employee of Barclays, Lehman Brothers and ISDA, another had been a senior executive at both Citi and Nomura and a third had been an executive at Bank of America and Paribas and was former CEO of ISDA. Thus, three of the four have been employed by at least one dealer member of the DC. On January 8, 2014 the DC met to consider External Reviewers for a determination on Caesars entertainment. They failed to publish the reviewable question that would be submitted to the panel, as required by the DC rules and announced that there are only four Reviewers (as opposed to the five required) in the Caesars pool, and that the chosen Reviewers are the same conflicted External Reviewers used in the CEMEX External Review.
This, coupled with the seeming inability to remove External Reviewers for undisclosed conflicts of interest, leads us to believe that the process of the External Review panel is a sham, that the DC doesn’t keep a list of potential Reviewers as required by rule and that they do not have a panel of five, also as required by the DC rules. In searching the web for other potential panelists who might have been selected, we were only able to find two, both of whom appear to have potentially serious conflicts of interests. One was a former COO at Goldman and continues to advise broker dealers and the other has regular engagements from Deutsche Bank.
Ending External Review
While the DC is expected to base its decisions on publicly available information, and all such information “shall be deemed to have been available to the Convened DC”, and the External Reviewers are only allowed to decide based on information available to the DC, the rules allow a simple majority to withdraw the Reviewable Question from the External Review panel if new information becomes available to them or the External Review panel is made aware of new information. This appears to be circular logic that allows a majority to end the panel if it appears the Reviewers are likely to vote against them. Given that the original DC process is run under the allowance that “neither the DC Secretary nor any DC Voting Member is under any obligation to research, investigate, supplement, or verify the veracity of, any information on which the relevant Convened DC bases its decision”, it seems that there is an inherent conflict in the rules. As a result, it seems that the only way a withdrawal could actually occur is if non-public information becomes available to either the panel or to the conflicted members of the DC, members who have no express prohibition – in the rules – against trading ahead or trading on non-public information.
Decision
In making a decision, the External Review panel, is once again biased in favor of the decision made by the majority of DC members. The ISDA rules state that unless the External Review panel is unanimous in their votes, and the original DC vote was less than the 80% supermajority required to render a decision but more than 60%, the External Reviewers decision automatically becomes the vote of the majority of the originally convened DC. In other words, as long as the majority can assure that they can get to a split decision among External Reviewers, which should be easy given the allowable financial conflicts among External Reviewers, they will have used the External Review process to give the appearance, rather than the reality, of a credible External Review.
How it Should Work:
As we previously stated, if the External Review process was rigorous, transparent, clear from conflicts of interest, and applied from the outset, as it should be, with respect to every reviewable question, it would render the biased, opaque and almost certainly manipulated Determinations Committee superfluous. Unfortunately, rather than being a fair process, imbued with independence and integrity, it appears to merely further the goals of that conflicted DC. As regulators and investigators investigate the self-regulatory structure of the CDS market, in which the largest players are able to control outcomes of CDS triggering events, regardless of the language in underlying debt contracts, we expect them to force necessary changes to the process and put in place proper regulatory oversight of the market. As we saw in the financial crisis, the manipulation of markets, by a concentrated handful of interested parties, undermines the integrity of those markets.
