The European Systemic Risk Board published last week an innocuously titled document called “Mitigating the procyclicality of margins and haircuts in derivatives markets and securities financing transactions“. The early pages, complete with jargon, hide an important call to action: introduce Initial and Variation Margin (IM and VM) for securities lending in place of current collateral practices. This could be called part 1 of a Standardised Approach to measuring counterparty credit risk, better known as SA-CCR, and create a new cost equivalency between OTC derivatives and securities finance transactions.
The SA-CCR is the Basel Committee’s rules for measuring exposure in the OTC derivatives market. It replaces the Current Exposure Method (CEM) and the Standardised Method, and calculates Exposure at Default (EAD) for each of five netting sets in the OTC derivatives market. These include FX for each currency pair, as well as interest rate, credit, equity and commodity derivatives. The US version has introduced some hybrid netting sets as well. The SA-CCR is meant to be a better measurement of risk for OTC derivatives than the previous methods. Importantly, it changes key model inputs on risk that impact cleared options and futures, leading to expanded capacity on balance sheets.
No such treatment currently exists in securities finance, and from a balance sheet perspective, a Total Return Swap beats a securities loan most every time for lowest cost. The greatest impact to securities finance comes in the counterparty exposure calculation as part of the Leverage Ratio. In a securities loan for example, a bank has exposure to both a beneficial owner and hedge fund, and in a cash collateral transaction also has a cash liability. There are also Liquidity Coverage Ratio impacts for securities finance transactions under 30 days, and Net Stable Funding Ratios for banks that already calculate this figure when transactions are for under one year. We first ran the numbers on the balance sheet differences between OTC derivatives and securities finance in “Regulatory Costs of OTC Derivatives vs. Securities Finance Transactions,” December 2015. Fast forwarding to 2020, it is little wonder that borrowing demand is down for the lowest value General Collateral securities.
The ESRB paper takes a different approach towards arguing in favor of IM and VM for securities finance:
The use of haircuts in SFT markets creates two types of risk. First, haircuts in non-centrally cleared SFT markets typically include counterparty-specific add-ons to mitigate counterparty credit risk. This can be a major source of cyclicality, as a deterioration in the perceived creditworthiness of the counterparty could trigger a generalised tendency to self-protect by raising haircuts. Second, while protecting the cash lender, haircuts expose the asset lender (cash borrower) to counterparty credit risk. Therefore, it is impossible to satisfy the need to reduce credit risk exposure for both counterparties at the same time by using haircuts.
Their proposal is to replace the collateral mechanism with IM and VM:
Replacing haircuts with initial margins paid to default remote entities would disconnect collateral payments from counterparty credit risk.
The ESRB does not explicitly call for an SA-CCR but rather employs the same principles to make the point. It also mentions the work already done in the OTC derivatives space to show how this would not be heavy-lifting for the for the securities finance industry:
A significant part of the infrastructure required to collect, value and segregate initial margins already exists, and a phasing-in period under such a proposal would enable market participants to customise this infrastructure for SFTs. Overall, this policy option would align non-centrally cleared SFT markets not only with centrally cleared SFT markets, but also with practices in non-centrally cleared derivatives markets.
A next step would be creating the same balance sheet accounting for securities finance under the ESRB plan as exists for OTC derivatives, or even preferencing securities finance for the simple reason that it helps market liquidity in physical instruments. (Thanks to the Division of Research and Statistics at the Federal Reserve Board for showing again how this works in “Over-the-Counter Market Liquidity and Securities Lending”.) It’s pretty well accepted that in capital markets, you don’t want the derivatives market driving pricing in the physical markets, and yet that can easily occur today. Rather, you want physical market price discovery leading the way for derivatives pricing. Driving business towards securities finance over Total Return Swaps, for example, helps promote this objective.
Firms reliant on cash collateral reinvestments won’t like a plan that takes away the cash pools in favor of margin exchanges. At the same time, non-cash markets have proven for years that effective pricing can be achieved on an intrinsic basis, and lower balance sheet costs for borrowers would go a very long way in amping up transaction volumes. CCPs have the same idea.
This is not the first call for SA-CCR changes with securities finance in mind, nor will it be the last. In May 2015, the RMA Committee on Securities Lending wrote to the Basel Committee on Banking Supervision asking for action:
The RMA Committee proposes that the SA-CCR be modified and adopted for agency securities lending transactions by mapping loan and collateral asset classes to one or more existing SA-CCR asset classes… Such a mapping would have the benefit of creating similar capital charges for transactions with similar economic exposures and would have the benefit of consistency across frameworks.
An SA-CCR for securities finance is a logical and somewhat obvious solution to the current capacity constraints in securities finance. The ESRB paper is concerned about risk for counterparties, not volumes or liquidity, but reaches a conclusion that the industry could support. An SA-CCR for securities finance would go a long way in solving multiple objectives that current regulations have derailed.