Something that missed our radar back in March was the Federal Reserve’s proposal to allow systemically important FMUs (financial market utilities) to establish accounts with the central bank and thereby get paid interest on their reserves, much like the primary dealers.
This sounds unsexy as it is, but the quick background here is that the Dodd-Frank bill empowered the Fed to supervise those FMUs that are designated systemically important by the Financial Stability Oversight Council. And along with the added supervision, those FMUs would be allowed to open the reserve accounts with the Fed.
A recent Citi note tries to quantify just how big those accounts would be if they were to open soon, and the possible impact on short-term rates.
But first an overview of those FMUs that have been designated systemically important and what they do:
1. Payment system – These are business [SIC] focused on netting and settling large value payments among a group of banks. CHIPS is a clearinghouse for interbank payments in the US and CLS is a clearinghouse for international, multicurrency payments.
2. Cash clearing – This type of FMU is primarily involved in netting and settling trades in cash securities. All of the designees of this type are subsidiaries of DTCC which provides “book-entry” ownership transfer services.
3. Central Counterparties (CCP) – Derivative CCPs are primarily engaged in clearing derivative transactions. For instance, if a client and a dealer clear a swap through CME, rather than facing each other, the client and dealer will both face the CCP. To manage counterparty risk, CME will demand that both parties to the swap post margin to the CCP.
We can ignore the payment systems here, as they don’t need to hold overnight balances of cash and securities.
But Citi estimates that right now the designated cash clearers and CCPs are holding roughly a combined $220bn in cash and securities as collateral. The CCPs account for most of it — with about $200bn — and of that amount the vast majority ($189bn) is in securities.
Here is how the analysts estimate how much of that $220bn would migrate to a Fed reserve account, and where it would come from:
Currently, FMUs can only access interest on Fed reserves indirectly, either by depositing cash in a bank account, lending to a bank through repo, or purchasing shares in a money fund that in turn funds banks by purchasing their CP and investing in repo. Indeed, CME reported that over $13 billion in margin it characterizes as “securities” is in fact cash invested in money funds through a program CME offers to provide clients with return on cash collateral.
If FMUs could directly receive 25bp interest on their reserves they would likely pull cash from repo and money funds, currently returning sub 25bp and open to operational and credit risk, and deposit the proceeds in their Fed accounts.
If we take CME as a benchmark, and assume about 15% of CCP securities are invested in money funds, this implies $28 billion in CCP MMF investment. This is on top of the $21 billion in FMU cash that is likely invested in repo markets, and any Tbill or MMF share collateral that is posted. Thus, the immediate impact of interest on reserves for FMUs would be a pull back of at least $50 billion from short-term markets.
That’s not much and would have only a modest impact on short rates, but the obvious point that the strategists go on to highlight is that the amount of collateral held by CCPs is likely to continue rising with the post-crisis regulatory push for OTC derivatives to be cleared. (See Lisa’s recent post for an overview.)
Estimates for just how much more collateral these institutions will ultimately hold, and how quickly, vary. But the point is that it will make the collateral available for repo and short-term lending markets even more scarce than it already is.
Giving CCPs the option to be paid on their reserves will help ease the crunch, as less money will be pursuing the extant collateral in money markets — and in doing so will also alleviate the downward pressure on short rates.
The establishment of paying IOR to banks in 2008 compressed secured and unsecured rates at a time when unsecured lending (and federal funds rate trading) had ceased. Bernanke himself said as much.
But recall that the expiration of the Transaction Account Guarantee this year amounted to a removal of collateral already, and also that the US Treasury just paid down some of its debt. Short rates are already under pressure, and will come under even more as derivative market participants begin posting more collateral.
Barring an unexpectedly fast acceleration in NGDP that lifts rates up off the zero lower bound generally, the Fed’s extension of IOR to new participants should also enhance its ability to maintain control over its transmission mechanism should the collateral scarcity worsen.
