Fixed-Rate Reverse-Repo Facility Update
Last month’s release of the July FOMC meeting minutes contained the words to describe the new Repo facility: “fixed-rate,” “full-allotment,” and “additions to the range of counterparties.” The meaning of those words in the context of Repo jumped out at me immediately. I speculated there was a major shift brewing in Fed overnight rate targeting, or even a complete overhaul of the tri-party system in the works. My view happened to be a little controversial, but it was picked up by a variety of other economists, including Credit Suisse’s “What’s happening in the Money Markets” on August 26. I even picked up a new moniker, “Chief Repo Watcher, Scott Skyrm” from FT Alphaville in an article they posted shortly thereafter. I like the new title, by the way. I’m looking for additional information and discussion about the facility with release of the September FOMC minutes on October 9.
Doom And Gloom About A U.S. Treasury Collateral Shortage
Honestly, if my “doom and gloom” warnings about a pending U.S. Treasury collateral shortage were correct, Treasury general collateral would be trading around .03% to .02% by now. Why did GC bottom out at 5 basis points in June and stay there? Given the estimates I ran for QE2 and QE3 + QE4, overnight rates should be close to zero by now. When the Fed continues to buy $85 billion a month, it’s taking collateral out of the market, and assuming no significant changes in cash, based on pure supply and demand, rates should go lower. What happened? New Treasury issuance hasn’t really changed, so that’s not it.
Could it be that Treasurys were dumped back into the market during the sell-off in the outright market that began at the end of May? That would put supply back into the Repo market. Given that 10 Year Note yields moved from 2.03% on May 22 to around 2.90% today, there’s going to be more supply of Treasurys that dealers need to finance. This could be part of the reason.
There’s another theory too. Perhaps there’s a minimum cost of doing business in the Repo market. Maybe dealers cannot drop overnight rates down to 0% and still get their clients to invest. Also, the GC Repo rate might be “sticky” once it gets to 5 basis points. When rates go that low, maybe some participants are priced out of the market. That is, they moved their cash elsewhere.
Clearly, the Fed is going to vote to “taper” at the FOMC meeting this week. First off, Fed officials have been preparing the market for a cut in bond buying since May 22. That’s standard operating procedure for the Fed when a change of policy is in the works. Second, at the Jackson Hole conference in August, a paper was presented by Robert Hall basically calling QE ineffective, though that view is clearly not shared by Ben Bernanke and Janet Yellen. They have supported QE and claim that it prevented the Great Recession from becoming worse. The paper said that QE1 was effective because it supported rates during a crisis, but subsequent programs, during a recovery, have not lived up to expectations. If QE was the “be all end all” of monetary policy, providing a free lunch to economies everywhere, then the paper critical of QE would not have been presented just one month before the planned tapering.
Next, if tapering is a given in September, then the next question is by how much? Sometimes I get mixed up between what I think the Fed should do and what the Fed will do. Of course, sometimes there’s a fine line between the two. The market consensus is for $10 billion or $15 billion of tapering, which I call worthless. When the Fed entered the realm of QE purchases, it was in $40 billion and $45 billion a month increments. If you take 25 basis points as the standard change in the fed funds target rate, then a $10 billion or $15 billion taper is like tightening 5 basis points at a time.
Specials – 10 Year Note
The August 10 Year Note (2.50% 8/23) clearly did not perform as well as the February and May 10 Year Notes in the Repo market. It actually makes sense though; that’s a typical trading pattern for Specials. There are basically three things to consider: First, with 10 year yields up around 3.00%, there’s a lot less fear of another major sell-off. At least, for now. Once the market has already backed-up, there’s less need to hedge and that means fewer shorts in the sector. Second, traders learned their lesson from the past two 10 Year Notes. “Shame me once, shame on you . . . shame me twice, shame on me.” Well, at least not a third time. Shorts will stay in older 10 Year Notes rather than roll forward. Third, traders will find other means to hedge instead shorting 10 Year Treasurys. Overall, I’d say the “bull market” in 10 Year Repo rates is over for a little while.
Remember all the hysteria about the European Financial Transaction Tax? Mine included. Well, never mind. A group of European Union lawyers concluded that the tax is illegal under the EU treaty. Hopefully, this will give the German and French governments cover to back out of continuing down the FTT road. The Italian government might notice their existing FTT is probably illegal too. However, the EU Tax Commissioner disagrees and believes the tax still has a legal footing. So, actually it’s not completely over as of yet.
Clearly, when regulators sound the alarm about the evils of re-hypothication, it sounds like one of those hidden specters of risk in the financial system. The recent FSB report in August raised concern about repo re-hypothication. Disclaimer: I have never seen nor read of any credible instance in which re-hypothication played a risk to the financial system. Let’s take a look at the mechanics of re-hypothication.
Re-hypothication means that securities can be borrowed and loaned from one counterparty to another, over and over. Take a U.S. Treasury 10 Year Note: A Securities Lender loans $100 million to a U.S. broker-dealer, who might lend it to a Canadian bank, who then lends the $100 million to a hedge fund. The 10 Year Notes get passed down the line in a chain of transactions. At first glance, it sounds bad if one link in the chain is broken, say one counterparty goes bankrupt. If that’s the case, how do the securities find their way from the hedge fund back to the Securities Lender? Actually, as I write it, it does sound very risky.
Here’s the good news – the legal framework of the Repo market provides for immediate liquidation of securities borrowed and loaned when one party defaults. If the Securities Lender loaned securities to the broker-dealer, and the broker-dealer went bust, the Securities lender can go into the market and buy back the loaned securities. That’s the first important point.
Second, we’re not talking about one-of-a-kind securities. Were talking about securities with billions outstanding. The Securities Lender needs to get back $100 million of the current 10 Year Note, not the exact same $100 million sitting at the hedge fund. There could be $40 billion worth of the U.S. Treasury 10 Year Note outstanding and they’re all identical. The key point is that there are billions of completely identical securities in the marketplace to be purchased.
Now what happens with smaller securities issues, like sovereign debt in Europe? Stock loan businesses have dealt with this issue for ages. In the bond market, when there are securities which are unique, with small amounts outstanding, they’re not loaned down the line through many counterparties. They’re generally funded by one cash provider in a tri-party or bi-lateral repo transaction. Perhaps moved to a second funding counterparty, but rarely to a third counterparty. That’s it. It’s a highly unlikely that a small bond issue with limited liquidity would be passed down the line so many times. And that’s why re-hypothication is not a major threat to the financial system.