Reducing Systemic Risk or Merely Transforming It?
Notes from the Vault
And so it begins. With the implementation of substantive Dodd-Frank rulemaking, so comes the first wave of innovation in response to those rules. We shouldn't be surprised, of course. This is just the latest example of the regulatory dialectic that Boston College Professor Edward J. Kane first described over 30 years ago. Adapted to the current situation, a financial crisis begets regulation, which begets innovation, which begets more regulation and perhaps, one day, another crisis, and on and on it goes. But this type of innovation can't really begin until the new regulations have been written, so that the regulatory and procedural gaps can be identified and exploited. The finalization and implementation of Dodd-Frank Title VII swaps regulation has created the incentive for innovation and the market is rising to the challenge. We have recently seen the creation of new securities (future/swap hybrids), but one of the most mentioned innovations in this space is a new financial service: collateral transformation. Assessing this new activity in the context of its effect on systemic risk provides a nice example of the regulatory dialectic at work, especially the tendency for innovation to expand the size of "shadowy banking."
Prior to the financial crisis, the vast majority of derivatives trading occurred in over-the-counter (OTC) markets rather than on exchanges. In these markets, trading involves two counterparties that can and do customize nearly every aspect of the contract being traded, including the collateral posted by each party and the rules by which that collateral needs to be adjusted over time and due to market conditions. In this way, OTC derivatives trading involved bilateral credit risk among counterparties where the amount and form of risk assumed was a function of the risk tolerances, objectives, and exposures.
Perhaps the most infamous securities that traded in this way were credit default swaps (CDS), termed "financial weapons of mass destruction" by Warren Buffett. During the crisis, calls for increased collateral from AIG's CDS counterparties were claimed to have worsened AIG's financial distress and precipitated its failure. To be sure, the presence of heterogeneous contracts and the unobservability of information about firm exposures exacerbated the transmission of financial distress and failure through a complex web of interconnected positions and counterparties.
To remedy what were seen to be regulatory gaps in the derivatives markets, Congress drafted legislation in the Dodd-Frank Act to standardize most derivative contracts and bring their trading and/or clearing onto exchanges (or swap execution facilities). Notably, markets for securities traded on exchanges, including derivatives such as futures and options, continued to function normally during the crisis and none of the exchanges suffered financial distress. So presumably, the thinking was that by bringing OTC derivatives into this framework, opacity would be reduced, allowing regulators and market participants to observe, monitor, and manage this activity better and reduce systemic risk. Additionally, the introduction of centralized counterparties (CCPs) would simplify and reorganize the interconnections between financial institutions and other market participants. As their name suggests, the CCPs become the counterparty to each side, both buyer and seller, of each trade. Thus, the CCPs assume the credit risk that used to be borne by the institutions engaged in an OTC trade. Reassigning this credit risk does not, of course, eliminate the systemic risk aspect of derivatives trading, but rather concentrates it in a small number of CCPs that can then be regulated with respect to risk management.
Just as in the prior OTC markets, CCPs manage the counterparty risk they take on largely through collateral requirements. A key difference is that collateral arrangements (including amounts, haircuts, and reconciliation frequencies) in the OTC market were customized by each set of counterparties according to their own assessments of the credit risk, risk tolerance, hedging options, and collateral management costs; in many cases, trades involved no collateralization. For example, the ISDA 2010 Margin Survey reports that 70 percent of all OTC trades were collateralized with 82 percent of these posting collateral received consisting of cash, 9.9 percent consisting of government securities, and 8.2 percent consisting of other securities. These overall rates partially reflect the differing characteristics of various contracts and the impact of these characteristics on the level of underlying credit risk exposure; that exposure is higher when contracts have long duration and when the volatility of the price of the underlying asset is larger. For instance, the collateralization rate for CDS trades was 93 percent while only 79 percent for fixed-income derivatives. Foreign exchange and commodity trades are collateralized at even lower rates and often employ letters of credit rather than asset collateral. When collateral was posted, it was often rehypothecated by counterparties, meaning that the institution that received collateral would in turn pledge this collateral as part of a different OTC derivative transaction. The resulting collateral trail added a further dimension to the complex transmission path by which financial distress could spread.
In contrast, CCPs will require collateral on all trades, impose a standard set of haircut and eligibility rules to all counterparties, and engage in daily reconciliation. The collateral requirements of CCPs are generally more narrow than pre-Dodd-Frank OTC practices, and they greatly incent the use of highly liquid securities with limited risk profiles, including cash and U.S. Treasuries, and render the collateral ineligible for rehypothecation (the exchange retains control of the collateral). Because of these differences, many market participants expect an increase in demand for high quality collateral as OTC contracts begin to migrate to CCPs. Estimates of the collateral required to meet initial margin requirements for formerly OTC derivatives trades at CCPs vary from $140 billion to $1.6 trillion.1 OTC derivatives trades that are not eligible to be cleared by CCPs will also now have collateral requirements under Dodd-Frank and estimates of the collateral that will be required to establish these trades ranges from $800 billion to $10.2 trillion.2 Additionally, the liquidity coverage ratio in Basel III, European Market Infrastructure Regulation (EMIR), and Solvency II are likely to further increase the demand for high quality assets globally among banks and insurance companies. (See this Bank for International Settlements report for an overview of the demand for collateral assets.)
The other innovation in this space is decidedly more complicated, at least in terms of its impact on systemic risk. "Collateral transformation" is often described by market participants as "upgrading collateral." By definition, however, unless you can somehow intervene and improve the creditworthiness of the issuing institution, upgrading collateral is not really possible. This practice is better described as swapping collateral (and indeed, one form of collateral transformation is a "collateral swap," which has traditionally been a bilateral OTC contract itself). The new wave of transformation services are being developed and marketed by dealers and custodian banks and utilize the repurchase (repo) market to "source" high quality assets for their collateral posting clients. Let's say a hedge fund only has BBB corporate bonds or equities to post as collateral. Unfortunately for the fund, these are not CCP eligible for the newly traded and cleared derivatives. The hedge fund can enter into a repo contract with a dealer or bank to transform this collateral into cash or U.S. Treasuries that are acceptable to post. The dealer/bank can serve as a direct counterparty in the repo or as an intermediary for trades with other institutions or funds that possess high quality assets or cash available for lending. As of now, at least some of these services would also allow rehypothecation of the lower quality assets.
It's easy to see what each side gets out of this transaction. The collateral-poster gets access to eligible collateral without having to liquidate current asset positions. The repo counterparty supplies high quality assets and earns a yield bump (especially if providing cash). And the dealer/bank earns fees for intermediation. In terms of the efficient allocation of high quality assets to the institutions that value them most, collateral transformation seems very efficient and beneficial.
What's less clear is whether the imposition of centralized trading/clearing, the concomitant collateral standards, and the resulting innovation of collateral transformation have, all told, significantly reduced systemic risk (see also University of Houston Professor Craig Pirrong's Streetwise Professor blog). The residual credit risk that was initially borne by OTC counterparties is now being borne in small part by the CCPs and in larger part by the banks/dealers and the repo counterparties. The opaque connections between OTC derivative counterparties have been eliminated, as each has the CCP as a counterparty, and replaced with what are likely fairly opaque repo market connections. Moreover, these connections are likely to involve entities like pension funds, insurance companies, and perhaps even nonfinancials, given the large amount of high quality and cash assets that these entities hold. Does this imply an increased potential for financial distress to be transmitted to these institutions and sectors? Perhaps.
The failure of an OTC counterparty had the potential to be a systemic event and, as mentioned, collateral calls were highlighted as contributing to financial distress at AIG. Collateral transformation does not fundamentally alter this scenario but could replace the collateral call with a "run" in the repo market, in which the suppliers of high quality collateral retreat from the market, forcing the liquidation of derivative positions and/or lower quality collateral instruments. This creates the potential for fire sales, which can contribute to systemic risk by quickly and substantially reducing asset values. Whether the resilience of the network of connections (i.e., its ability to withstand institutional failures) will increase or decrease is also unclear. What seems somewhat more predictable is that more crucial financial activity will be channeled through, or involve, dealer firms and custodian banks. This raises the ante on their regulatory oversight and may increase systemic risk via concentration effects.
Thus, derivative markets are in the first of possibly several rounds of Kane's regulatory dialectic in which (re)regulation is followed by innovation around the regulation. Whether Title VII and its associated regulations prove to be beneficial, harmful, or merely benign in the next crisis remains to be seen.
Paula Tkac is a vice president and senior economist at the Atlanta Fed. The views expressed here are the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System. If you wish to comment on this post, please e-mail firstname.lastname@example.org. You can also read previous Notes from the Vault posts.
1IMF 2012 Global Financial Stability Report, Chapter 3, BIS "Collateral Requirements for Mandatory Central Clearing of Over-the-Counter Derivatives," March 2012.
2ISDA "Initial Margin for Non-Centrally Cleared Swaps: Understanding the Systemic Implications," 2012.