The OFR recently issued its annual financial stability report in which it found that “the financial system is now far more resilient than it was at the dawn of the financial crisis” but that potential threats to financial stability still exist. The report identifies three such threats, one of which is resolution risks at global systemically important institutions (GSIBs) – and, in particular, the continued risk posed by large derivatives portfolios. In our view, the report offers an uninformed view of how large bank resolution works, and so we attempt to correct the record here.
The key issue at hand is what happens to the derivatives portfolio of a large banking organization if the organization fails and enters into resolution. The report correctly notes that, historically, derivatives portfolios have created significant challenges when large firms, particularly nonbank firms, fail – Long-Term Capital Management in 1998 and Lehman Brothers in 2008 are two good examples.
The challenges stem mainly from the fact that bankruptcy law allows derivatives counterparties to terminate their contracts with the failed nonbank firm immediately, instead of subjecting them to a temporary stay, the way both the bank resolution law and Title II of the Dodd-Frank Act do. And because they can run, derivatives counterparties oftentimes do run by terminating and closing-out their contracts, seizing collateral (further draining precious remaining liquidity out of the failed firm) and liquidating it quickly (which may result in a fire sale, depressing prices for key widely-held asset classes, which can lead to contagion).
The report (again, correctly) notes that early termination problems generally don’t arise when insured banks fail, because they are resolved under the Federal Deposit Insurance Act, which has a temporary statutory stay on the termination of derivatives contracts. Dodd-Frank’s Orderly Liquidation Authority (OLA) also has a temporary statutory stay that can be used, but of course only if OLA is invoked. These temporary stays are critical because they provide needed time to transfer derivative contracts to a bridge institution or other solvent counterparties that can continue performance, thereby avoiding the problems caused by early terminations.
The lack of a temporary statutory stay in the bankruptcy code is ameliorated by the fact that GSIBs are incorporating a temporary contractual stay in all of their financial contracts, including derivatives, which will apply in bankruptcy. This is being done pursuant to regulatory mandate for GSIBs to amend their financial contracts in a manner consistent with the ISDA resolution stay protocol, which imposes a two-day stay on counterparty termination rights and continues these stays if certain conditions are satisfied. The OFR report, to its credit, takes note of the ISDA protocol.
But then the OFR report veers off track and provides a surprisingly uninformed analysis of how derivatives portfolios of GSIBs would be treated in a resolution under bankruptcy. This happens primarily because the report fails to correctly understand how the temporary stays work in tandem with the single-point-of-entry (SPOE) recapitalization strategy, which is the strategy adopted by 7 out of the 8 US GSIBs in their latest living will plans, and is expected to be pursued by all 8 US GSIBs in their next plan submissions.
Under these plans, which outline how the firms would be resolved under bankruptcy, the losses of a troubled GSIB would be pushed up to its top tier holding company, which would fail and file for bankruptcy, imposing losses on or wiping out its equity and debt holders. The failed GSIB’s critical operating subsidiaries, holding nearly the entirety of the GSIB’s derivative portfolio, would be recapitalized and provided with liquidity to meet their ongoing obligations. These operating subsidiaries, newly free of losses and freshly recapitalized, immediately would be transferred to a new largely debt-free holding company, along with any parent guarantees of their financial contracts. Alternatively, the newly recapitalized subsidiaries would continue to operate outside of bankruptcy under the bankrupt parent, and guarantee claims would be elevated to administrative priority status in the parent’s bankruptcy case.
This would all happen over “resolution weekend,” well within the stay periods under the ISDA protocol. As a result, the GSIB’s derivatives counterparties, including counterparties to its nonbank affiliates, would be unable to exercise termination rights unless there were a direct default under the derivatives contract. Rather, the counterparties would face operating subsidiaries that have been freshly recapitalized and provided with liquidity, and therefore able to meet performance obligations, even if just for long enough to pursue an orderly unwind of the derivatives portfolio, which is required to be described in detail in the resolution plans.
The OFR report recognizes none of this, and instead counterfactually asserts the failed GSIB’s derivatives portfolio would need to be transferred to a third party rather than staying with the operating subsidiaries. Transfers to third parties would be necessary, the report states, if the nonbank affiliate operating subsidiaries “need official sector liquidity support during bankruptcy” or were otherwise unable to meet their contractual obligations, e.g., variation margin payments. In other words, the report worries the operating subsidiaries will not have the resources or financial strength to carry on after resolution weekend.
The report’s analysis is flawed, and ignores the fact that GSIBs, as part of their resolution planning, include estimates of the capital and liquidity resources needed by their operating subsidiaries to pursue an orderly unwind of their derivatives portfolios, post-resolution weekend. These estimates are made pursuant to the RCAP/RCEN and RLAP/RLEN models required by regulators in the resolution plans. The models are subject to rigorous testing and validation procedures under direct regulatory oversight. Furthermore, the GSIBs have structured their resolution plans, including via triggers for the holding company’s bankruptcy filing as well as through binding secured support agreements, to ensure these resources will be available at the operating subsidiaries as needed.
Thus, the report misses the mark on the treatment of large derivatives portfolios of GSIBs in resolution. By misunderstanding how they are treated, and by not appreciating the plans in place to ensure they are dealt with in an orderly and well-resourced way, the OFR finds uncertainty and risks that do not exist in fact.
The report makes a couple of other flawed assertions worthy of correcting. First, the report faults regulators for requiring only GSIBs to modify their derivatives termination rights, leaving other banks and nonfinancial firms out of scope. But this fails to recognize that the regulators have affirmatively stated they would monitor the market and expand the scope of firms required to adopt the ISDA protocol beyond GSIBs as warranted. In addition, the size and complexity of a bank’s derivatives portfolio is one factor the regulators evaluate in determining GSIB status.
Second, the report cites as a weakness the fact that the ISDA protocol does not apply to centrally cleared derivatives, and therefore CCPs could terminate contracts with GSIBs upon an affiliate’s filing for bankruptcy. While true the protocol doesn’t apply to cleared derivatives, the report fails to discuss whether any CCP has the incentive to terminate given the GSIBs’ SPOE recapitalization strategies, or whether the CCP’s regulators would even allow it. The report also fails to note that there are initiatives under way at the Financial Stability Board and elsewhere to address GSIB continuity of access to CCPs.
Taking a step back, it is important to acknowledge that the new resolution frameworks for GSIBs, like many other post-crisis reforms, have yet to be fully tested in an actual future crisis, so certainly some risks remain. However, by failing adequately take note of the full scope of the reforms enacted and how they would be operationalized, the OFR appears to dramatically overstate the risk of GSIB resolution on both an absolute and relative basis. To the latter, we see numerous systemic risks — for example, a continuing divergence between fixed income issuance volumes and fixed income liquidity — that should have been ranked far higher than the one on which the OFR focused.
 Transfer to third parties, the report goes on to suggest, would face a number of uncertainties and present a variety of risks, some dubious, some not. For example, would two days be enough time to do a transfer to a buyer? Is any single firm big enough to assume an entire portfolio? If not, how easy is it to break up the portfolio?
 RCAP/RCEN are shorthand for “Resolution Capital Adequacy and Positioning” and “Resolution Capital Execution Need,” and generally refer to amount of total loss absorbing capacity, including capital and unsecured long-term debt, of a GSIB, as well the pre-positioning of recapitalization resources within the firm to ensure that material operating subsidiaries have the resources needed to allow them to operate or be wound down in an orderly manner following the parent company’s bankruptcy. RLAP/RLEN are shorthand for “Resolution Liquidity Adequacy and Positioning” and “Resolution Liquidity Execution Need,” and generally refer to the overall liquidity needs and resources of a GSIB, including a methodology and plan for ensuring liquidity is available after the parent’s bankruptcy filing to stabilize the surviving material entities and to allow those entities to operate post-filing.
Disclaimer: The views expressed in this post are those of the author(s) and do not necessarily reflect the position of The Clearing House or its membership, and are not intended to be, and should not be construed as, legal advice of any kind.