The GC-IOER collapse is far more significant than many people realize. The dynamics of the repo market are not always easy to understand, but somehow I don’t think that is the reason it has received zero attention outside of ZeroHedge.
ZeroHedge’s coverage has rightly focused on the liquidity implications for Wall Street institutions and the overall repo market. There is also the push to get money funds out of repo and in to the regulated banking system through deposits. While these are certainly significant issues in and of themselves, I think there is an arguably more substantial aspect that has not yet been uncovered.
In Tyler’s last post on the subject, May 18, he quoted Barclays Joseph Abate:
“We believe this will increase the premium to general collateral at which these issues trade and make it harder to cover shorts.” [emphasis added]
For those unfamiliar with the repo market, it has a number of essential functions. Chief among them is its mechanism to cover short positions in fixed income securities. If the special collateral rate is negative, as the May 18 post indicated it was for a third of all UST repos, then the moneylenders that are short on UST positions cannot find enough physical bonds to cover them. Shortage of this magnitude is often a function of hypothecation.
The answer to the counterintuitive question of why someone would lend money at negative interest rates is that the moneylender’s first priority is to cover that short position. As the special rate deepens in the negative or persists there for a period of time, the pressure on the moneylender to cover grows exponentially.
This type of negative collateral lending rate is currently on full display in the silver market, where forward rates have persisted at negative levels since April 21 (before that in February & March). Silver forward rates are nothing more than another form of collateralized loan, very much akin to special collateral repo trades. When forward rates are negative, the demand to cover short physical positions forces the moneylender to pay out interest because they need to cover more than they need to earn a positive rate.
The longer the negative interest penalty for being short remains, the more pressure there is to exit the short position. It is a self-reinforcing feedback loop since the short covering forces prices higher, leading to more losses for holding short positions, etc.
From the perspective of the Federal Reserve, what better way to “influence” interest rates on the longer end of the yield curve, particularly if rates are trending in the “wrong” direction. There is no evidence that the Fed colluded with the FDIC to change the GC-IOER carry dynamics, but a look at interest rates since early April shows a happy Fed.
Tyler’s first post, dated April 5, coincided with a top in five-year treasury yields. In that post Tyler reported a sharp downturn in repo rates, with GC hitting 6 bps. He followed that up with a post on April 12 where GC rates fell to the ridiculous 1 bp level. Since special rates are below GC rates, we can safely conclude that most securities “on special” were at negative rates.
Again, “coincidentally”, the ten and thirty-year treasuries both topped out the day before the GC rate fell to that low level. Yields across the longer end have been falling ever since. Could this simply be the Fed pushing rates lower through forced short covering in the repo market?
The Fed doesn’t mind playing with fire (in this case liquidity) since it’s history is littered with ill-conceived interventions. We know there is pretty strong circumstantial evidence that the Fed is concurrently trying to influence rates through volatility (see ZeroHedge http://www.zerohedge.com/article/did-fed-its-stealthy-synthetic-bet-keep-yields-low-become-next-aig).
We also know that any decrease in demand for treasury shorts would negatively impact treasury auctions. This seems counterintuitive, but shorts do their work in on-the-run securities. In 2002, the Federal Reserve studied this phenomena and concluded:
“Dealers sell short on-the-run Treasuries in order to hedge the interest rate risk in other securities. Having sold short, the dealers must acquire the securities via reverse repurchase agreements and deliver them to the purchasers. Thus, an increase in hedging demand by dealers translates into an increase in the demand to acquire the on-the-run security (that is, specific collateral) in the repo market.” i
The demand for the on-the-run securities by short traders keeps demand high at auction. So, the more treasury shorts there are, the greater demand at auction time (we studied this in March 2010 and concluded this would be one of the prime factors keeping rates low after QE 1.0 ended. Everyone in the world was convinced rates would move far higher, yet they never did as auction after auction showed increasing and bewildering strength – the demand for shorts was keeping the auctions well subscribed).
Conversely, in April and May 2011, the forced short coverings have led to less than enthusiastic auctions. Again, from the Fed’s perspective, these are slightly negative indications for interest rates, but it more than makes up for them by running the shorts out of the repo market.
I would like nothing more than to offer conclusive proof that this is the case. However, absent any smoking gun, there is certainly enough coincidences to warrant giving this more thought. Treasury rates may be declining on their own (I personally doubt it), but given the timing of these moves it is not hard to imagine that the Fed and its regulatory cousins gave treasury rates a powerful shove in the “right” direction
i “Special Repo Rates: An Introduction”, Mark Fisher. Federal Reserve Bank of Atlanta, Economic Review, Second Quarter 2002, Page 27.