By Tyler Durden at Zero Hedge
This is a big deal. On the heels of our pointing out the surge in Treasury fails (following extensive detailing of the market’s massive collateral shortage at the hands of the unmerciful Fed’s buying programs), various ‘strategists’ wrote thinly-veiled attempts to calm market concerns that the repo market (the glur that holds risk assets together) was FUBAR. Even the Fed itself sent missives opining that their cunning Reverse-Repo facility would solve the problems and everyone should go back to the important business of BTFATHing… They are wrong – all of them – as yet again the Fed shows its ignorance of how the world works (just as it did in 2007/8 with the same shadow markets). As JPMorgan warns (not some tin-foil-hat-wearing blogger with an ax to grind) “the Fed’s reverse repo facility does little to alleviate the UST scarcity induced by the Federal Reserves’ QE programs coupled with a declining government deficit.” The end result, they note, is “higher susceptibility of the repo market to collateral shortages” and thus dramatically higher financial fragility – the opposite of what the Fed ‘hopes’ for.
Reverse repos do little to alleviate UST collateral shortage
This week’s FOMC minutes provided important hints about the Fed’s exit strategy. The interest on excess reserve (IOER) rate will play a “central role” during the policy normalization process while the overnight reverse repo facility (ON RRP) would play a “supporting role” by establishing a soft floor for money market interest rates. The Committee expressed concerns not only about the potential size of the reverse repo facility, which grew rapidly since testing began last September, but also about conducting monetary policy operations with non-traditional counterparties.
Our colleagues Alex Roever and Mike Feroli have written extensively about the reverse repo facility and its impact on US money and rate markets. In this note we focus on the issue of collateral shortage and try to answer a rather narrow question: Are reverse repos alleviating collateral shortage? [Spoiler Alert: NO!]
In principle, the reverse repo facility reduces UST collateral scarcity as the Fed sells a security to an eligible RRP counterparty, thus supplying UST collateral to the market, and at the same time it drains “reserves” from the financial system which are replaced with reverse repos in the liability side of the Federal Reserve’s balance sheet. It is important to emphasize that this reserve drainage does not alter the overall liquidity injected by the Fed. It merely diverts this liquidity away from banks to non-banks, including money market funds and GSEs. But in terms of collateral shortage alleviation, we see limited improvement for several reasons:
1) The reverse repo facility has been growing rapidly but at between $100bn-$200bn currently is rather small compared to the $2.6tr of excess reserves in the liability side of the Fed’s balance sheet or the $4tr of securities in the asset side, $2.4tr of which are USTs.
2) This week’s minutes, coupled with the FOMC setting a still wide 20bp spread between the IOER and ON RRP rates, suggests that the Fed has little appetite to allow the reverse repo facility to grow to very high levels eventually. The minutes specifically mentioned that “participants discussed several potential unintended consequences of using such a facility and design features that could help to mitigate these consequences” and “a number of participants expressed concern about conducting monetary policy operations with non-traditional counterparties.”
3) The USTs sold to counterparties via the reverse repo facility are in the triparty system. Tri-party repo is a transaction for which post-trade processing is outsourced by the parties to a third-party agent. This ensures transaction efficiency and better mobilization of collateral but it does not change the legal relationship between the repo parties. With the ON RRP confined to the US tri-party system, the USTs sold to ON RRP counterparties cannot become available to other counterparties that are outside the tri-party system such as hedge funds and/or small/medium sized asset managers.
4) In addition the tri-party system applies to general collateral repo transactions so potential high demand for specific USTs, i.e. “specials”, will be difficult to be satisfied. This is because the USTs sold to ON RRP counterparties can only be re-used within the tri-party system and under general collateral repo transactions.
5) Higher margin requirements as a result of recent regulations on OTC derivative markets, for example, have caused a rise in collateral demand. But securities held within the tri-party system in the US are typically not allowed to be used to satisfy margin requirements. This means that the USTs released via the Fed’s ON RRP facility will not have the same effect in alleviating increased collateral demand stemming from higher margin requirements, than if the Fed had directly sold these UST securities to open markets.
6) Eligible RRP counterparties are currently 139 covering a wide range of entities—94 of the largest 2a-7 money market funds, six governmentsponsored enterprises (Fannie Mae, Freddie Mac, and four Federal Home Loan Banks), 18 banks, and the 21 primary dealers. That is, the Fed offers Treasury securities via its reverse repo facility to a wide set of counterparties including both banks and non-banks. But all these institutions together do not account for more than a quarter of the overnight tri-party volume. In other words, the current set of counterparties captures a modest share of the tri-party market.
7) The RRPs are expected to be collateralized by Treasury securities. SOMA’s holdings of agency debentures and agency mortgage-backed securities are available for use, but are not expected to be used in this exercise. That is, reverse repos have the potential to alleviate UST collateral scarcity but not agency collateral scarcity. Admittedly substitutability within the broader government collateral universe should reduce the importance of this last argument.
In all, we believe that the Fed’s reverse repo facility does little to alleviate the UST scarcity induced by the Federal Reserves’ QE programs coupled with a declining government deficit. And the still rising trend in UST repo fails since QE3 started in Sep 2012 suggests that the issue of UST collateral shortage remains. It is true that the repo fails are focused around specific issues, particularly hot run treasuries. It is also true that at times and in various tenors the fails spike as the desire to short USTs by certain investors exceeds the floating stock of these specific issues. But what is more concerning is the rising trend in fails which can be seen in Figure 2. The picture in Figure 2 lends support to the idea that the leverage ratio regulation has permanently reduced the appetite of broker-dealers to engage in repo markets. Via its reverse repo facility, the Fed is effectively facilitating the withdrawal of broker dealers from repo markets.
Similar to what happened in the US corporate bond market, the end result is not only structurally lower repo turnover (Figure 3) but also higher susceptibility of the repo market to collateral shortages (higher frequency of spikes in repo fails as shown in Figure 2).
* * *
But why do I care about some archaic money-market malarkey? Simple, Without collateral to fund repo, there is no repo; without repo, there is no leveraged positioning in financial markets; without leverage and the constant hypothecation there is nothing to maintain the stock market’s exuberance (as we are already seeing in JPY and bonds).
Crucially, it should be inherently obvious to everyone that the moves we see in the stock market is not about mom and pop choosing to invest in the stock market (or not) as the ‘cash on the slidelines’ fallacy is “completely idiotic’ but about the marginal leveraged machine (or human) quickly jumping on momentum.
The spike in “fails to deliver” highlights a major growing problem in the repo markets that provide that leverage… and thus the glue that holds stock markets together.
Wondering why JPY and bond yields have diverged so notably from stocks in recent days… repo effects (it’s just a matter of time before it hits stocks)…
So that explains why the Fed is so desperate to talk you into selling your bonds – most notably the short-end by demanding you listen to what Yellen said about raising rates… as that reduces the shortfall of collateral that repo needs and restocks the banks with repo-able funds.
* * *
Is that why a noted dove like Jim Bullard was so visibly hawkish last week? The irony of course of the Fed explaining how rates will rise faster is that it spooks stock investors who have grown used to exuberant liquidity supply and rotates them to bonds… which merely exacerbates the problem the Fed has.