Over the past month we have been closely documenting a major funding squeeze in the all important shadow economy - the "synthetic liquidity" conduit which far more than traditional sources of cash, has become all important for proper bank functioning over the past decade. Courtesy of adverse development in Europe, one by one various components of this unregulated funding scheme have become frozen necessitating the first of many central bank interventions on November 30 to provide liquidity to global banks, primarily to offset such shadow conduits as locked up commercial paper, repo and money markets. Logically, as noted over a week ago, European banks scrambled to obtain cheap dollars by borrowing over $50 billion from the Fed, and plug dollar shortfalls. Yet as all band aid measures designed to offset a broken liquidity equilibrium fail eventually, it was only a matter of time before we saw a direct bail out by the Fed of one or more banks in the aftermath of the November 30 global "bail out." Sure enough, we have our first clue that "something" happened in the week ending Wednesday December 14 that involved an upgrade of the Fed's indirect (and thus untargeted) bailout of global banks, to a focused, and very much targeted rescue of one (or more) banks. And with some additional diligence, it may be possible to narrow down the date of an actual bank bailout: Tuesday, December 13.
Exhibit A - Reserve Bank Credit
Two years ago, when discussing the transition of the world to one coordinated, centrally-planned regime we said that the only financial statement of any importance, updated weekly, is the Fed's H.4.1, or the "Factors Affecting Reserve Balances" which traces that flow of "last resort" cash from the Fed to the various organization that make up the reserve bank, primary dealer, and various other financial entities under the Fed's Lender of Last Resort umbrella. Simply said, anything abnormal in this weekly report of "flow and stock" (a simplistic distinction where the Fed is far more focused on what the absolute level of reserve numbers is, whereas Zero Hedge and the market believe it is the "flow", or marginal change, that determines, artificially, asset prices) would confirm our speculations that the Fed has stepped into into its now traditional role of bailing out the world.
The first thing that caught our attention was the all important total reserve bank credit - the most important big picture metric announced by the Fed on a weekly basis. As the chart below shows, after having plateaued with the End of QE2, and remaining stable during the duration of the "sterilized" Operation Twist (as it should), in the week ended December 14, total reserve credit soared by a whopping $81 billion or the most since May 27, 2009 when the Fed was actively undergoing the early stages of QE1 damage control.
So what was the reason for this huge jump in reserve credit? Two things - on one hand we had the already long-ago telegraphed increase in Fed liquidity swap lines by over $50 billion, or from $2.3 to $54.3 billion to be exact. However that does not explain the remainder. So where did the other $30 billion in credit expansion come from?
Exhibit B - The Plot Thickens: First Net MBS Bulk Purchase Since QE1
It appears that in addition to reverting to such an "old school" QE1 global bailout mechanism as FX swap lines, the Fed also did something it had not done in a long time, or since QE1 to be exact: it bought a boatload of Mortgage Backed Securities, an act it last engaged in on a net basis back on August 11, 2010, which in turn was a delayed settlement of an earlier purchase. As a reminder, the Fed's balance sheet settles any MBS purchases on the mid-month update so while the big spikes in the chart below between January and July 2010 are indicative of broad MBS purchases by the Fed under the auspices of QE1, when it was out purchasing a total of $1.25 trillion in MBS in hopes of lowering mortgage rates and stimulating housing, and thus employment (something it failed at miserably), in the mid-month week just ended, the Fed bought, and settled concurrently, an unprecedented $31 billion in MBS.
Obviously $31 billion jump in settled MBS purchases is notable considering the pattern of previous MBS net flows since August 2010. But under what auspices did the Fed go ahead and buy this whopping amount of Mortgage debt? And why?
As New York Fed itself tells us:
Agency MBS Tentative Purchase Amounts and Historical Operational Results
The Desk’s tentative agency MBS purchase amounts associated with the reinvestment of principal payments from agency debt and agency MBS in agency MBS are shown in the table below. The numbers listed are subject to change, should the Federal Open Market Committee (FOMC) choose to alter its guidance to the Open Market Trading Desk (the Desk) at the New York Fed during the monthly period or if market conditions warrant. The amounts listed are approximately equal to the amount of principal payments from agency debt and agency MBS expected to be received over the monthly period, adjusted for any variations from prior periods, as described more fully in the FAQs.
In addition, in order to ensure the transparency of its agency MBS transactions, the Desk will publish historical operational results, including information on the transaction prices in individual operations, at the end of each monthly period.
Specifically, in the period between December 13 and January 12, the Fed had permission to buy, wait for it, $30 billion.
And yet, there is a discrepancy as in a subpage detailing gross and net purchases the Fed reveals only $7.550 billion in net MBS purchases for the week ended December 14.
So obviously while the temporal matching is still not precisely clear, what is clear is that in the week ended Wednesday, The Fed provided a whopping $81 billion in additional reserve credit between FX swaps and MBS purchases, the latter having no other purpose than to release even more liquidity to banks which have simply converted one illiquid security, into another: cash. This answers the important question of "why" the Fed did what it did. It is also unclear whether this outlier transaction was demand driven or forced by the Fed. All that will be confirmed once we get the official breakdown of MBS POMO on January 13. Incidentally, here is what typical MBS purchases and sales look like on a monthly basis (excel table).
While these two balance sheet outliers would have in themselves made for curious observations, if insufficient to draw any particular conclusions, it is Exhibit C that puts things into perspective.
Exhibit C - Average Discount Window Borrowings
When the Fed updates its H.4.1 every Thursday at 4:30 pm, it provides two sets of data: an average over the period, and a period end number. And if one was looking to find a flashing red light within Bernanke's book, which has always without fail been a big change in Discount Window borrowings (either Primary, Secondary or Seasonal Facility), looking at the period end number would have shown nothing out of the ordinary: there was a modest $42 million borrowed from the Primary Credit Discount Window facility on the day ending December 14, Wednesday, far less than previous 2011 outliers. However, things rapidly change when one observes the average usage of the Discount Window for the past week.
The result is as follows: a $393 million surge in average borrowings:
And since we know that of the 7 days that make up the average period, one can be eliminated (as there was no borrowings on Wednesday), the implication is that on one day in the week ended December 14, the Fed lent out up to a whopping $2.5 billion (as the $393mm is an average 6 day number) to a bank in the form of last recourse cash via the Discount Window.
Confirming just this speculation is Barclays' Joseph Abate:
After months of virtually no use of the Fed’s discount window, borrowing jumped to an average of $400m/day in the week through Wednesday. The Fed reports only the weekly average of daily borrowing and the daily amount outstanding on Wednesday. From these figures, we estimate that on one day last week, total discount window borrowing reached $2.5bn. Of course, the same $400m/day weekly average could have been achieved with a bank borrowing $900mn on Friday. It is unclear what prompted this pick-up in borrowing from the Fed. There was neither a spike in the fed funds rate nor any disruption in the repo market, so we are a bit puzzled. Of course, under Dodd-Frank, the borrowing bank’s name will be released – after two years.
Yes, the name of the bank that received what amounts to a Fed bailout will be released in two years, but no, we disagree that there was no disruption in the Repo Market. Perhaps Joseph forgets that the Fed lends out Discount Window cash to "eligible" entities out of Europe... where the repo market is in total collapse and wholesale disarray.
Furthermore, the borrowing was from the Primary Credit facility, or that reserved for stable banks, not Secondary Facility eligible names which have to pay an addition 50 bps in punitive interest. And since the bulk of Primary Credit eligible banks domestically already are swimming in $1.6 trillion in fungible excess reserves (which is the reason why discount window borrowings have been so modest ever since QE1 unleashed a liquidity tsunami for the bank, which serves no other reason than to plug capital shortfalls - it certainly is not being lent out) it is obvious that the Fed is now back to its old job of bailing out banks. And not just any banks - European banks.
Some appropriate reminders from the Fed on the "lender of last resort" discount window use:
|Above the FOMC's target for the federal funds rate.||Primary credit rate plus 50 basis points|
|Overnight||Short-term, usually overnight. Can be extended for a longer term if such credit would facilitate a timely return to reliance on market funding or an orderly resolution of a failing institution, subject to statutory requirements (FDICIA restrictions).|
|Depository institutions in generally sound financial condition; same as eligibility for daylight credit.||Depository institutions that do not qualify for primary credit.|
|Generally no restrictions.|
May be used to fund sales of federal funds.
|As a backup source of funding on a very short-term basis, or to facilitate an orderly resolution of serious financial difficulties.|
|Ordinarily no questions asked.||Reserve Banks will collect information necessary to confirm that borrowing is consistent with regulatory requirements.|
Depository institutions to which the law grants access to the Discount Window and which the Federal Reserve deems generally sound are eligible to obtain primary credit. Reserve Banks determine eligibility on an ongoing basis using supervisory ratings and capitalization data; supplementary information, when available, may also be used.
(CAMELS or equivalent)
1, 2, or 3
Adequately or well capitalized
4 or 5
Less than adequately capitalized
Common Borrowing Situations
The new Discount Window programs offer an enhanced opportunity for eligible depository institutions seeking an efficient solution to meet unexpected, short- term funding needs.
|Likely Situations for Borrowing Primary Credit|
Generally, there are no restrictions on borrowers’ use of primary credit. Here are some examples of common borrowing situations:
- Tight money markets or undue market volatility
- Preventing an overnight overdraft
- Meeting a need for backup funding, including a short-term liquidity demand that may arise from unexpected deposit withdrawals or a spike in loan demand
- Arbitrage opportunities
We know two things with certainty: In the week ended December 13 (14th excluded) one or more banks, most likely European, borrowed up to $2.5 billion from the Fed's Primary Credit Discount Window. And since US banks are drowning in dollar-based liquidity, any need to approach the Discount Window now, in the context of trillions of Excess Reserves, carries with its exponentially greater stigmata than it ever did during Lehman days. Also, in the week ended December 14, the Fed did a mid-month settlement of $31 billion in MBS purchases - a transaction which allowed a Primary Dealer to source critical liquidity, based on $30 billion in buyback authorization granted for the period beginning December 13. What we do not know for fact is whether the $30 billion in MBS purchases was completed on Tursday or Wednesday, and whether this is a delayed settlement for previous purchases, although due to the mid-month settlement process, it is possible that any transaction could have settled immediately. And for those seeking a specific "bank bailout" date, the 13th looks quite reasonable: it was the first day when an MBS purchase was permitted and it was the last day when a bulk Discount Window loan could have been performed.
But wouldn't the market learn of even a hushed European bailout? And wouldn't there be a massive sell off if it became clear that exactly two weeks after the Fed's coordinated broad bailout of European banks, it had to engage in another, far more politically tenuous bailout, this time via a $2.5 billion free money loan to a cash scrambling bank? Well, if the news was leaked at 2pm on Tuesday it sure would explain the market reaction...
So while much of the presented above is circumstantial, perhaps the next time Congress is debating with Ben Bernanke just how good it is for the US taxpayers to bail out European banks, someone can ask him just who it was that the Fed once again bailed out the week of December 14. Because America obviously does not have enough problems of its own...