Reverse Repo Test #3 Now A Fact: $225 Million In Liquidity Sopped Up By Fed

Tyler Durden's picture

Just a day after the second reverse repo test was conducted, the Fed has launched repo test #3: this time, as we expected, for a greater (if still notionally meaningless) amount of $225 million, compared to yesterday's $180 million. The term of this operation is just one day, compared to the 8 and 3 in tests #2 and #1. The increasing frequency of these Temporary Market Operations should be making the liquidity bulls very nervous. And as we pointed out yesterday, the notional on the repo test "can only go up" - so far we have been proven right. Yet wait for the collateral to move down from Treasurys. That's when all hell will break loose.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Anonymous's picture

This is one of those concepts that leaves me in a bit of a fog. Can anyone more well versed in these matters give a bit of an explanation (or link to one) of the mechanisms of reverse repos and their subsequent effects on liquidity/balance sheets? Why would increasing frequency (or, as I read it, decreasing duration) make liquidity bulls nervous?

I'll hang up and listen.

Anonymous's picture

In a repo, the Fed lends money to the dealers in return for various collateral. This means the dealers have cash which they "spend" and it gets into the banking system Thus the Fed is supplying liquidity to the financial system.
A reverse repo works the other way. The dealers lend cash to the Fed in return for collateal. Thus cash is taken out of the financial system, thereby reducing liquidity
The term of these loans is generally short term; typically overnight or for a few days, but can be for a longer period. I think that a term of more than a month would be rare.
Banks and dealers also arrange repos (and reverse ropos) amongst themselves. These will only be for their liquidity needs and have no impact on general financial liquidity from the Fed's perspective.

bobby02's picture

"Yet wait for the collateral to move down from Treasurys. That's when all hell will break loose. "

That's the 2nd time in 2 days you have written that. Why? If the counterparty makes good on the repo, the collateral is irrelevant. Since we know the primary dealers are tbtf and will be saved, counterparty risk is zero.

(If counterparties at a later data move beyond the primary deals, then this may become an issue.)

Also, if we take you argument that the collateral will be shit, then the Fed is in the cat bird seat, since if the reverse transaction does not take place, the Fed has the cash and the counterparty the shit collateral, presumably with a minimal haircut (i.e. Fed gets more $)


SayTabserb's picture

Just a guess, but I think a reverse repo involving the Fed's MBS portfolio causes hell to break loose because any transaction which sheds light on the actual value of this stuff is very dangerous for the entire financial system. The actual "deal" of these toy transactions, insofar as they affect the parties involved, doesn't really matter.

Anonymous's picture


Wake me up when they get over a billion in one shot.

Anonymous's picture

it doesn't matter if Fed uses FNMA mbs or UST ..... the rate paid simply changes a few bps .

The 'liquidity fear' at work here is the Fed tries to drain alot of cash out of the system quietly as a means of reducing QE without actually taking the ' club to head ' approach of raising Fed Funds ( rate hike ) .

The $64 question is even if they take this up to a few hundred billion , will that really change short rates ? or better put , at what level of reverse repo do we see Fed Funds structure begin moving higher due to cash becoming scarce .

I would suggest all here read Crudele's story in NY Post the other day . He tracks the BLS birth/death model carefully . He says BLS has now figured out that they added in 824,000 new jobs they simply assume were created but in fact were not . BLS does a one time huge 'death' adjustment every January ( see the peak in 2009 was in january of 2009 ) .

if Crudele is right , January unemployment drop with 824,000 subtracted out will blow Street out of water with their 'miss' and make the testing of liquidity drain in these repo's meaningless .....

Assetman's picture

I cannot imagine they would test more than $5 million of MBS collateral-- if any at all.  The issue is that you start coming up with market clearing prices on the collateral when you do these deals.  Then more pressure will be placed on the Fed to determine how much they paid for this junk in the first place.

Reverse repo or not, if I were a primary dealer, I'd stay the heck away from the MBS colleteral deals.

Cursive's picture

I can't square this either.  This trade cannot be unwound.  In fact, I never thought of it as a trade so much as a perpetual government backstop.

Jim ODonnell's picture

This is just a ploy by the FED to act as if QE is over, not so. Here is what I posted on this subject Dec 4, 2009 at 3:30 AM EST. So far it has worked, gold and most of the rest of the inflation trade has been chrushed.

chinaguy's picture

Suppose the Fed (TBTF) marks the MBSs to 100% on the dollar, and they are traded to other TBTF banks, and bought back by the TBTF Fed. WHERE is the mark-to-market price disclosure that everyone is so worried about? I don't see where the problem (for the Fed & the other TBTF players) is in this, especially as the PDs have been treated so well in the past. 

What am I missing that is salient?


demsco's picture

The reverse repo is for show only. QE is going nowhere. Interest rates are going nowhere. They are doing this to shut people up about the government funding huge bank profits. Now they simply can say, see we are trying to drain the system! We all know they cannot really do a meaningful liquidity drain, probably ever, or raise interest rates to a meaningful level, ever. If they did the collateral is dead and the debt servicing costs will be sky high before 10 years are up.