In the past two weeks, Liberty Street Economics published two papers describing the current state of the federal funds market. The cash side is dominated by the Federal Home Loan Banks (FHLBs) and the borrowing side is dominated by foreign banks. Then, just last week, a senior Fed official described how the combination of the Fixed-Rate Reverse-Repo facility (FRRP) and paying Interest on Excess Reserves (IOER) can be used in concert to better control overnight interest rates. Let me do a quick caveman summary: fed funds bad, FRRP and IOER good. Is the Fed setting up to move away from the dying fed funds market and use a new interest rate policy tool?
The State of Fed Funds
The fed funds market is an overnight market for non-collateralized bank lending. It includes banks, thrifts and the GSEs as the major market participants. The size of the fed funds market has shrunk considerably since the beginning of the financial crisis – from trading $200 billion a day in 2007 to about $60 billion a day in 2012. What’s the reason for the dramatic drop in volume? Clearly, it’s the IOER. When the Fed started paying an above market rate on excess bank reserves, cash naturally moved straight to the Fed. However, not all fed funds market participants have access to the Fed. The FHLBs are left out and they’re stuck receiving regular fed funds market rates while other banking institutions can deposit cash directly with the Fed. According to Liberty Street Economics, banking institutions make up 26% of cash lending in the fed funds market, with the FHLBs making up the rest. On the borrowing side, on average in 2012, foreign banks borrowed about $30 billion a day and account of 42% of the borrowing volume. Shockingly, much of the foreign bank volume is assumed to be an arbitrage between the fed funds market and the IOER – borrowing cash in the funds market and depositing it at the Fed, and earning the spread in the process.
Look at it this way: almost 50% of the fed funds market is comprised of a U.S. government entity loaning cash to foreign banks which in turn loans that cash to the Fed. The fed funds market has evolved into a kind of arbitrage opportunity between government agencies. At the same time, the fed funds market is still the Fed’s primary interest rate for setting monetary policy. I’m sure I’m not the only one wondering how long a $60 billion market – about the size of a large Treasury Cash Management bill and being primarily used for arbitrage – can continue to be the Fed’s benchmark interest rate.
Talk of The Fed Lowering The IOER
Back in October and November, there was talk about the possibility of the Fed cutting the IOER. First, it was implied in the release of the October FOMC minutes. Then, the Minneapolis Fed President said in a speech on November 12 that the Fed might consider dropping the rate. Even the ECB has discussed cutting their equivalent of the IOER to a rate of -.10%, which could possibly open the door for the Fed to follow.
The basis of cutting the IOER is that banks are depositing too much of their cash at the Fed instead of making loans. Since Fed started paying interest of excess reserves, over $2 trillion in cash has been deposited at the Fed. However, there’s a large downside in cutting the IOER. Imagine what would happen if $2 trillion suddenly flooded into the overnight money markets. Massive cash moving into fed funds, GC Repo, and T-bills would push all of those rates straight down to zero. It would effectively destroy short-term interest rates, which would then spill over into the money market funds, destroying that industry in the process.
Some economists argued that a cut in the IOER is more likely because of the FRRP. With the FRRP setting a floor of .05% in the overnight GC Repo market, it potentially allows the Fed to cut the IOER rate. However, there’s a serious flaw here. The amount of securities available at the FRRP is only $139 billion ($1 billion per counterparty with 139 counterparties) and there’s over $2 trillion in cash sitting at the Fed as excess reserves. If the IOER were cut, there would be a tsunami of cash wiping clear through the FRRP, pushing rates straight down to zero.
I don’t think the Fed will cut the IOER, but my reasoning is not based on the recent strength in the economy. I think something else is happening:
Putting It All Together
Last week, Simon Potter, the NY Fed’s Executive Vice President and manager of the System Open Market Account (SOMA) said that the FRRP “offers improved control of the level of money market rates and reduced volatility of short-term interest rates,”* and it should be used along with the ability to pay interest on excess reserves – the IOER. I see the FRRP as having set a floor in overnight interest rates and the IOER effectively setting a ceiling.
The IOER “ceiling” was last hit in September 2012 when massive net Treasury issuance was still hitting the market each month and there was no QE program in place. Though there was pressure for overnight GC Repo rates to move higher, they were anchored by the IOER rate set at .25%**
So, putting it all together: Liberty Street Economics (The Fed) says the fed funds market is small and ineffective. The manager of the SOMA program said the Fed now has tools to set an upper and lower bound for overnight rates. Since the fed funds rate is becoming irrelevant, and the new short-term interest rate policy tool could be a rate band between the FRRP and IOER. Picture the Fed moving the FRRP rate up to .15%, it would, under normal market circumstances, keep the GC Repo between .15% and .25% each day. A rate band clearly gives the Fed the ability to set overnight market rates exactly where they want them.
* Market News Service; December 4, 2013, “NY Fed Offl: O/N Fixed Full-Alltmt Rev Repo ‘Promising'”
** I assume the transaction costs for a bank to move cash from IOER to GC Repo would be about 5 basis points, that’s why rates never really moved about .30%