Fixed Rate Reverse-Repo (FRRP)
I would describe the Fed’s Fixed-Rate Reverse-Repo program as very successful. The FRRP has eliminated the chance of a collateral shortage. It’s a mechanism to dump up to $139 billion in Treasury collateral into the market and put a floor on GC Repo rates. Clearly, it has protected the general collateral market (GC) from flat-lining at 0%, or dipping into the negatives; an otherwise unintended consequence of QE. Last week, the Fed increased the FRRP rate from 2 basis points to 3 and then raised it again to 4 basis points effective November 12. Ultimately, the goal appears to be 5 basis points, the maximum allowed by the FOMC, at the moment. That brings up the question as to whether the FOMC intends to raise the rate higher than 5 basis points at some point in the future. I don’t think they will. Unless the FOMC has the intention of using the FRRP as a larger tool in monetary policy, then 5 points is an appropriate floor in Repo rates and to prevent collateral shortages.
It will be interesting to see to impact on the FRRP facility as Repo rates moves higher, especially during quarter-end and year-end. If Repo rates are at .25%, will anyone show up at the Fed for collateral? They did during the debt ceiling crisis, but that’s an exception. So far, it’s clear the FRRP program is also a “window dressing” facility. From recent data, use of the facility doesn’t just spike on quarter-end, but also on month-end. It makes sense that banks and other financial institutions want the Federal Reserve as their Repo counterparty for statement periods. I expect the FRRP program to become even more popular on year-end.
Over the past month, the FRRP wasn’t the only thing putting collateral into the Repo market. GC Repo rates have averaged slightly higher since the end of the crisis. For the three weeks prior to the crisis, GC averaged 6.7 basis points, compared to 8.4 basis points for the three weeks after the crisis. Just like after the July/August 2011 debt ceiling crisis, collateral came back into the market. When end-buyers are selling to the dealers, more securities enter the Repo market. In 2011, it solved the collateral shortage problem created by QE2. Remember, just as QE2 ended and the Repo market was teetering on a collateral shortage in July 2011, Treasurys were dumped into the market due to fears of a debt default. The difference in 2011 was that the Fed’s buying program had ended, so the possibility of a collateral shortage was clearly over. This time around, continued buying to the tune of $85 billion a month will eventually drive GC rates down close to zero again. Luckily, this time, the FRRP is in place.
Actually, the October 2013 debt ceiling crisis isn’t really over. It was just kicked down the road into next year. Under the agreement, the debt ceiling reaches its limit on February 7, 2014, no matter what number it is. On that date, the U.S. government is not allowed to borrow anymore. Of course, that’s just when the borrowing authority runs out. Wrightson estimates that there’s about $200 billion in “extraordinary measures” which will, at a minimum, take us to March 13. That’s the date of the next possible technical default.
Repo Market Regulation
The Fed’s Mark Van Der Weide said the Fed is “looking at potential margin requirements for securities financing transactions.” Just like the FSB announced in the UK last August, there is discussion about standardizing haircuts for securities financing. In my view, this would be a problem to implement. Unlike something like futures margin, where customer cash and securities deposits are held at the exchange or FCM, how do you standardize Repo margin? In the Repo market, both cash and securities can flow both ways – from bank to customer and vice versa. Is the collateral seller or collateral buyer required to post the margin? Suppose a hedge fund is lending a bond to a bank? Now, suppose the transaction is the reverse? Do you standardize the haircuts based on dealer to customer? But then the market ends up with all kinds of complicated formulas, like Basel III ended up so much more complicated than Basel I.
Liquidity Coverage Ratio
It’s also been reported that the Fed is looking at stricter LCR rules than required by Basel III. On top of that, the Fed wants to implement them one year earlier than required by Basel III, supposedly with 80% fulfilled by Jan 1, 2015. That’s just a little more than one year away. The purpose of the LCR is for banks to hold enough high quality liquid assets to be able to sell them in case their funding is cutoff during a crisis. The target is to fund the bank’s operations for 30 days.
Getting back to the Repo market, it’s estimated by the Fed, no less, that banks will need to come up with an additional $200 billion in high-quality assets to comply with the new rule in 14 months. The LCR could ultimately increase high quality collateral demand by $2 trillion. The question then becomes – where does all of this collateral come from? There’s already a fear of high quality collateral shortages due to new regulation and the OTC derivatives migration to exchanges.
There’s a report that the Fed is probing banks on their exposure to REITs, specifically those that hold mortgage-backed securities (MBS). It’s estimated that REITs held $160 billion in MBS at end of 2009 and those holdings shot up to $460 billion in March of 2013.* Ironically, the Fed is looking into the effects of a bond market bubble, a bubble they created by juicing up the market up with QE purchases.
The problem with REITs, as least as the Fed sees it, is they buy MBS (long-term assets) and fund them in the Repo market on a short-term basis, generally from month-to-month. A typical shadow banking function. If there’s a significant sell-off in the long-end, like we saw between June and September, REITs could be stuck with large losses and forced to sell their positions, thus, creating fire-sale risk. Those losses lead to more selling which leads to more losses and the circle continues. It’s a hot topic in regulation these days.
The National Association of REITs shot back in an advertisement in the Wall Street Journal today. Though they admitted that REITs, as a whole, lost 2.7% in the third quarter, their point is that it’s not significant – it certainly doesn’t mean there’s a flaw in their business model. The NAREIT also cited that between June 2005 and June 2006 interest rates rose substantially and REITs posted a gain of over 20%.
Keep in mind, first, that most major REITs are public companies and must disclose any mark-to-market losses. Second, no matter where the long-end of the bond market goes, the fed funds rate is still set at 0%. If market expectations are reasonably correct, it means that short-term funding costs are anchored near 0% well into 2015. The real worry for REITs should not be as much a long-end sell-off, instead it’s when the Fed starts hiking short-term rates. That’s when the cost of funding could exceed interest income if short-term rates move higher fast enough.
* Financial Times; “Fed probes banks’ exposure to mortgage vehicles”; October 27, 2013