Back in the day, when the sophisticated deep thinkers who effuse deep economic thought, were deeply contemplating whether the Fed's IOER was a better tool to assist if and when (hint: never) the Fed begins to hike rates, or whether the relatively new (conceived about a year ago) Reverse Repo was the better candidate to help push liquidity out of America's bloated financial institutions, we made it very clear that the entire debate is completely irrelevant, as the only purpose of the Reverse Repo was to assist banks in pretending (with the Fed's explicit knowledge) that they have a better balance sheet than they represent.
We did this first in January in "Window Dressing On, Window Dressing Off... Amounting To $140 Billion In Two Days", [13]then in April in "Month-End Window Dressing Sends Fed Reverse Repo Usage To $208 Billion: Second Highest Ever [14]", then in June "WTF Chart Of The Day: "Holy $340 Billion In Quarter-End Window Dressing, Batman [15]", then in July "Record $189 Billion Injected Into Market From "Window Dressing" Reverse Repo Unwind [16]."
Of course, the abovementioned deep thinkers ignored this because it would mean that all the argumentation about the Reverse Repo facility as a means to assist the rate hiking cycle was irrelevant, and that instead of hiking rates the Fed was far more concerned with the collateral shortage that the TBAC loudly complained about in the summer of 2013... just months before the RRP was unleashed (recall "Desperately Seeking $11.2 Trillion In Collateral [17]"). Pure coincidence, right?
Well, the argument largely became moot when two weeks ago, following the Fed's recent announcement Reverse Repo would be capped at $300 billion [18], leading even the deepest of pundits to realize they have been fooled all along, and the RRP facility was never meant to be a rate hike-facilitating mechanism, the Fed released this: [19]
Statement to Revise the Time of Day of the Overnight Reverse Repurchase Agreement Operation for September 30, 2014
As noted in the September 17, 2014, Statement to Revise the Terms of the Overnight Reverse Repurchase Agreement Operational Exercise, the Open Market Trading Desk (the Desk) at the Federal Reserve Bank of New York (New York Fed) has been working internally and with market participants on operational aspects of tri-party reverse repurchase agreements (RRPs) to ensure that this tool will be ready to support the monetary policy objectives of the Federal Open Market Committee (FOMC).
Regarding the operation to be conducted on Tuesday, September 30, 2014, the Desk will conduct the operation several hours earlier than usual, from 8:00 to 8:30 a.m. (Eastern Time). All other terms of the exercise will remain the same.
This change only applies to the operation conducted on September 30, 2014. The operations conducted from Monday, September 22, to Monday, September 29, and those conducted on and after Wednesday, October 1, will be conducted at the previous time of 12:45 to 1:15 p.m. Any future changes to these operations will be announced with at least one business day’s prior notice on the New York Fed’s website.
Wait, why did the Fed explicitly warn that just one reverse repo operation would be temporally-adjusted, namely that of September 30, i.e., today? Simple: what is today?
Why quarter end, "window dressing" day of course.
In other words, the Fed ripped off the mask that RRP was anything more than a way for the Fed to allow banks to appear more palatable to... drumroll... Fed regulators. Regulators such as Carmen Segarra, who once again made the news [20], not only for being fired for daring to ask probing questions about the Fed's "close" relationship with Goldman Sachs, but for providing 48 hours of recording confirming that the NY Fed is merely a branch of Goldman Sachs.
So fast forward to today at 8:30 am when the Fed announced [21]the result of today's "special" window-dressing Reverse Repo operation. What was unveiled blew our socks right off, because not only was the Reverse Repo an absolutely whopping $407 billion, but the low rate on the auction was an unprecedented -0.20%!
[22]
And here is what today's operation looks like in historical context:
[23]
That's right: at $407 billion, it far exceeded the $300 billion new cap on the program.
So yes: everyone can now admit that Reverse Repo was nothing more than Fed-mandated window dressing, no point in covering that up any more.
But what about that Low rate of -0.20%?
For the answer we go to Stone McCarthy which has done a forensic analysis of precisely what happened in today's "Crazy" (as they call it) Reverse Repo operation.
From SMRA [24] [24]:
Quarter-End ON RRP Craziness
Today's quarter-end ON RRP offering from the Fed included a total of $407.167 bln in submissions, far exceeding the new $300 bln overall cap on the program
As such the 5 bps fixed rate was not applicable, and the allocation was decided by an auction mechanism. Bidders (since September 22) are required to include in their submission a rate at which they would be willing to engage in the Fed's RRP operation. Today these bids ranged from a low of -20 bps to a high of +5 bps.
The Fed by starting at the lowest rate and working upward was able to do the $300 bln max at a stop-out rate of 0%. All awards were at this stop-out rate. Thus the bidder at -20 bps was probably envisioning a stop out rate well above their -20 bps bid.
Today's offering was a test of the program. The Fed conducted this operation at 8:30am this morning in anticipation of quarter-end considerations. Typically the dealine for the operation is 1:15pm. The stop-out rate was even lower than what we envisioned. We were thinking 1 or 2 bps.
What this means is that today's offering was done at a rate below the fixed rate of 5 bps, and thus today the floor aspect of the ON RRP program was not effective. This will probably be the situation at most future quarter-end offerings.
Oops: this means that the RRP as a mechanism to hike rates will certainly fail due to the discontinuity of collateral requirements, which spike at quarter end. Because try as it might, the Fed simply has no way of hiking rates on all other days except March 31, June 30, September 30 and December 31.
Why such a surge in submissions?
As quarter-end approaches dealers typically pare positions for balance sheet dressing purposes. They may also be less willing to engage in matched-book RP activities in helping financing their customers.
The lull in dealer financing demand means that the MMFs, the primary counterparty party to the Fed offerings, have liquidity to put to work that is redirected to the Fed ON RRP offering.
At previous quarter-ends the MMFs have accounted for around 89% of the "take-up" of the ON RRP offerings.
This compares to around 82% on non-quarter-end dates.
Does this argue for a higher cap than the $300 bln?
Not necessarily. The MMFs, of course, wish that cap were higher. If so, such may have provided ample quarter-end investment opportunities with positive interest rates. Indeed, some Treasury bills were trading at negative interest rates in response to the capping of the ON RRP program at $300 bln, previously there wasn't a cap.
The cap was imposed because the FOMC was worried that in times of financial distress (not routine quarter-ends) the MMFs would only engage in lending to the Fed with the dealers and other money market borrowers getting cut off.
Some worried that the Fed might become too dominate a player in the money markets.
The results of today's offering are not really surprising. The Desk probably anticipated something close to what happened here. We think that what we saw in today's offering will be typical of future quarter-ends. Despite that fact that the MMFs would probably prefer a much higher cap, thereby rendering somewhat higher quarter-end returns, what they are earning (0%) is still better than the negative returns on bills that mature early in the new quarter. In other words, the ON RRP program is still a better alternative than what would exist in the absence of this program.
And there you have it.
A bigger problem, however, emerges, now that it is empirically proven that the Reverse Repo is now meaningless and doomed as a means to allow the Fed to hike rates in a world in which the Fed Funds rates is irrelevant, and a parallel rate corridor somehow has to be established. Which means that only the IOER fallback exists, a rate hike environment fallback which as we wrote back in 2012 is also meaningless as it only controls for one half of the rate increase corridor.
So... still betting on a rate hike in mid-2015, when the Fed itself has just admitted, and the market has confirmed, it has no clue how it will hike rates?
We'll take the much, much over.
