Readers may recall that a week or two ago, there was a good deal of chatter in the ether about reports issued by the Federal Reserve that showed that the banking system, for the first time ever recorded, showed negative net non borrowed reserves.
Some commentators and analysts were alarmed; others regarded it as no big deal, arguing that those figures had little meaning. Our position was somewhere in the middle. We noted, as others did, that the change resulted from the implementation of the Fed’s Term Auction Facility. We argued that whether this change was a cause for concern depended on why banks were using the TAF so heavily. If it was merely because the funding was cheap (which it is, lower than interbank rates) then this is simply rational profit seeking. But if it is because the banks need the dough and might have difficult raising the funds otherwise, then the TAF use indicated continuing distress.
To give a quick overview of the TAF: it was launched December 17, with two $20 billion actions, one for 28 days (the one conducted on the 17th) and a second for the 20th for 35 days. The reason for the program was that the gap between the Fed funds rate and interbank rates had become very large, suggesting that banks were reluctant to lend to each other. That was even more acute in December, since banks customarily curtail their short term lending then so they can tidy up their books for year end.
We’ve called the TAF a discount window without stigma (and in fact, the Fed implemented the TAF because banks weren’t using the discount window even when they should have). Banks can post a wide range of collateral, borrow on a non-disclosed basis, and can hold on to the cash for a while (by contrast, the discount window is overnight)
Only two auctions were initially announced, but it was assumed that if this device worked, it would be continued for an indefinite time, presumably till the credit markets calmed down.
So what happened? The spread between interbank rates and the Fed funds rate fell, the TAF has been deemed a success, and some have even called it a lasting feature in the Fed’s arsenal.
But are things all that rosy? The Financial Times today raises some concerns, noting that banks are indeed using the TAF to use crappy collateral for borrowing,
And note that, with no announcement I can recall, the facility has been increased to $50 billion even though the year end crunch has passed. That too is not a good sign.
From the Financial Times:
US banks have been quietly borrowing massive amounts of money from the Federal Reserve in recent weeks by using a new measure the Fed introduced two months ago to help ease the credit crunch.
The use of the Fed’s Term Auction Facility, which allows banks to borrow at relatively attractive rates against a wider range of their assets than previously permitted, saw borrowing of nearly $50bn of one-month funds from the Fed by mid-February.
US officials say the trend shows that financial authorities have become far more adept at channelling liquidity into the banking system to alleviate financial stress, after failing to calm money markets last year.
However, the move has sparked unease among some analysts about the stress developing in opaque corners of the US banking system and the banks’ growing reliance on indirect forms of government support.
“The TAF . . . allows the banks to borrow money against all sort of dodgy collateral,” says Christopher Wood, analyst at CLSA. “The banks are increasingly giving the Fed the garbage collateral nobody else wants to take . . . [this] suggests a perilous condition for America’s banking system.”
The Fed announced the TAF tool on December 12 as part of a co-ordinated package of measures unveiled by leading western central banks to calm money markets.
The measure marks a distinct break from past US policy. Before its introduction, banks either had to raise money in the open market or use the so-called “discount window” for emergencies. However, last year many banks refused to use the discount window, even though they found it hard to raise funds in the market, because it was associated with the stigma of bank failure.
The Fed has not yet indicated how long the TAF will remain in place.
But the popularity of the scheme is prompting speculation the reform will stay in place as long as the financial stresses last.
“Some Fed officials have expressed an interest in keeping and possibly expanding the TAF,” says Michael Feroli, economist at JPMorgan.
Nevertheless, Mr Feroli said banks now appeared to be using the TAF instead of other funding routes, meaning that the overall level of reserves in the system was remaining constant. “The banking system certainly has its problems, however the notion that . . . banks have trouble maintaining reserves stems from a superficial reading of the Fed’s statistical reports,” he said.
Caroline Baum did a story on this, too. Basically saying that this negative reserves thing really ain’t no big deal. And she’s exactly right, taken on it’s own, it ain’t no big deal. Now if there were a severe credit crisis going on simultaneously, or if major institutions were regularly reporting multi billion dollar losses, or the markets were weakening, or the reserve currency of the world were looking particularly weak, or if gold/silver/oil/wheat/soy/etc were all busting out to all time highs… those sorts of things might convince me that this negative reserves thingy might be a big deal.
The TAF are similar to the long term refinancing operations of the European Central Bank. However the long term operations done by the ECB are larger (currently 270 billions of euros) and longer (3 months).
So I don’t understand why the FT says it is a bad thing for the US and good thing for the Europe (Trichet has been elected man of the year by the FT).
I am not as familiar with the ECB’s operations as I ought to be. One thing I do know is that even when they have appeared to inject a lot of liquidity, they have mopped it up quickly too. Dunno how that plays in here.
The big compliant in the article is that the banks are posting particularly crappy collateral (and what it doesn’t mention is that they are getting below market rates, when if you believe Bagehot, the banks should pay a penalty rate).
Is the ECB’s rate below interbank rates? And how stringent are they in the sort of collateral that has to be posted? If you could shed some light here, that would be helpful
The TAF will always be fully utilized. After all, it’s an auction. There will always be a price level at which everything is taken, unless there is a very serious freeze in the banking world.
As the TAF will not increase the amount of money long term, assuming that it is not continuously enlarged, it is a great way for the Fed to gain insight into the natural demand of money out there.
It would be nice, if the Fed gave up on trying to fix the interest rates at all, and would instead move entirely into rates being set by the markets. Then they could focus on controlling the actual money supply.
The Fed can gain good insight into which banks are in the most trouble, by looking at their bidding levels at the term auction facility.
Bad collateral gives the banks more incentives to make high bids to roll over their TAF loans if they are in trouble.
By accepting bad collateral, the Fed has gotten the banks into a hook. Once a troubled bank is hooked on the TAF, they can not easily get away from it. When they run into serious trouble, the Fed can close it down or arrange a take-over.
Negative non-borrowed reserves are a function of TAF. The reserve position is irrelevant in and of itself.
TAF is a facility that has been encouraged by the Fed. As long as its offered, banks will take advantage of it at a rate that makes sense.
Obviously the banking system is still in distress by various measures and events. The TAF is a Fed insurance policy to avoid the risks of withdrawing it at this time.
So worrying about the implications of either the negative non-borrowed reserves or the continuation of TAF per se is pointless. These are symptoms of a problem and its risks. You might as well worry about why the Fed has dropped rates so much, or why some projections of bank writedowns are so extreme, or why Shedlock thinks the world is coming to an end because of non-borrowed reserves.
The ECB accepts ABS with a “single A” credit assessment as collateral (The implementation of monetary policy in the euro area, September 2006). I have also read that the ECB accepts also the tranches of CDO with the same rating.
After the crisis of August, the ECB has modified the allocation. It has increased the long term refinancing operations and it has reduced the short term operations. The total amount has only slightly increased (perhaps an increase of 20 billions).
TAF is currently running at $60 billion, two auctions of $30B each, issued on 31st Jan and 14th Feb respectively. SOMA holdings have been reduced by a similar amount in parallel with this.
We have a novel facility that the Fed launched to deal with extreme conditions in the credit markets. Even though many commentators attributed at least some of the reduction in banks’ reluctance to lend to each other to normal end of year activity, the Fed has increased the size of the TAF operations by 50% as Anon of 8:16 tells us since they were launched in December, rather than decreased it, as one might expect.
That is symptomatic of continued stress in the financial system. I am at a bit of a loss to understand why some readers take exception to that observation.
“That is symptomatic of continued stress in the financial system. I am at a bit of a loss to understand why some readers take exception to that observation.”
Perhaps they work for the FED, Goldman, or just a party hack. They want to calm the waters.
“I am at a bit of a loss to understand why some readers take exception to that observation.”
Because continued stress was what the Fed was allowing by sheduling further auctions to being with. And the reserve effect is a complete red herring. Let’s not double and triple count the intensity of market stress by adding up indicators that are just reflections of one another. It’s bad enough as is. And Libor spreads have declined significantly on balance.
As Anon earlier points outs, the Fed is doing 2 tranches of $30B each month in 2008 ; they upped it from $20B tranches and did announce this on Jan 4th
I take issue with the reporter’s suggestion that “dodgy collateral” is posted – it wouldn’t surprise me but its conjecture at this point; the TAF results press releases gives no indication of the type of collateral posted; we don’t even get told of the high low marks of the rates, like we do on TOMO operations, that would let us infer somewhat about the collateral posted.
So, did the reporter just make a fact up as he/she went along ?
seems the issue here is transparency like everything else. The Fed has essentially taken what wa a transparent process and made it opaque – for the greater goog of course. Can you imagine if a compny did such a thing? where is Spitzer with his FD mantra. We have a problem so less obfuscate and pretend that everything is ok. Then lets slash interest rates, lend them money and try and recapitalize them before the end comes. This is not only unamerican, it is crinimal. barrons has a good take on the return to a gold standard in theis weeks magazine. The better FT article would have been a call for the auction participants to be listed so the market can make its own diagnosis. Kind of reminds you of the monline exposure issues. If the great free markets advocates would come out of their holes and call for transparency we could get on with the purge. We don’t need a new sheriff, we need a Chapt 11 reorg — and fast
Newspapers have fact-checkers and standards for reporting. The FT is based in the UK, which has much tougher libel laws than the US, so they probably run a tight ship.
They might have gotten several comments from people who wanted to stay off the record.
Banks are not merely posting dodgy collateral, they are creating new collateral specifically for the purpose of posting it (with the Fed or the ECB). Collateral which no one else would be interested (or perhaps, able) to buy.
“The big compliant (sic) in the article is that the banks are posting particularly crappy collateral (and what it doesn’t mention is that they are getting below market rates, when if you believe Bagehot, the banks should pay a penalty rate).”
The TAF bidding starts at the relevant OIS swap price, i.e. where the fed effective overnight fixing compounds up to for the duration of the relevant period. The ois (overnight index swap) is basically the markets’ true expectation of Fed rates going forward. Therefore I would argue that getting money from TAF at the lowest possible rate is NOT ‘below market rates’ – but bang on market. The rates that the TAF money comes below (currently, but not not necessarily depending on demand) are the Libor rates…. but the whole point of the TAF (in my view) was to get the Libor rates down and to get the interbank cash market moving again!
All that said, I agree that these TAF auctions or ECB equivalent (3 month LTRO’s) have started to produce the (unwanted/unforseen?) effect of banks creating RMBS and keeping them on the banks books as a quick way, in another liquidity squeeze like August, to access ECB/FED liquidity. Which in itself is simply proper foresight and liquidity management – but at the end of the day it all points to a need for further re-capitalisation of banks or/and decreasing balance sheets.
I have a few questions arising from the article above, I wonder if someone will be able to enlighten me!
The FT article and some comments suggest that the Fed (and other CBs) are accepting lower quality collateral. I presume in the event of a borrower defauting then the central bank would take ownership of the collateral. If this happened, what would the effect on the defaulting bank be? Could it survive? And what would the central bank do if the collateral turned out to be worth less than its lendings? Does it just print more money to cover its losses or would it need to sell other securities or assets to cover them. Another point I am unsure about is to what extent can a central bank accept poor quality collateral, would there come a point when the currency markets would lose faith in the currency maintaing its value and thus act to put a limit on the central bank’s ability to accept poor collateral? Do central banks have to report what is on their books like regular banks do, and if not how do the markets form an accurate picture of the health of an economy?
Thanks for any help!
What happens with this TAF money if the bank breaks? does it belong then to the taxpayers?
Actually, Yves, what was worrying was not that “dodgy” collateral might be accepted by the TAF, in layman’s terms the TAF is the federal equivalent of the “we accept all collateral” pawnshop , so “dodgy” collateral is to be expected. What is really worrying is that the TAF operation has signs of becoming a semi-permanent feature that as the article said ” lasts as long as the stress lasts”. Perhaps it aims to solve the medium term financing problem? Just the very thought of that should raise dismay amongst those who hoped for a quick crash and quick recovery. BTW, Credit Suisse’s explanation for their discovery of “pricing mistakes” on a routine review in the process of a financing issue ought to send up a red light, god knoiws who else is gonna find more mistakes!
I have not been sufficiently clear here. According to a theory regularly invoked by some serious economists, set forward by Walter Bagehot, in times of stress, central banks should lend against good collateral at penalty rates.
The notion is if you don’t, you create moral hazard. The penalty rate is to punish the banks for having managed its affairs badly; the good collateral requirements is similarly to avoid rewarding them if they make stupid risky bad loans.
So here we have, first, a facility that was supposed to be short-term the Fed now seems to regard as permanent. I’d feel better if they were signaling that they intend to cut it back at some point in the future. Second, it is getting bigger even though conditions could have been expected to improve early in the year, thus lessening the need for it. Third, if what 2:53 says is correct, the TAF is distorting behavior, which is consistent with what Bagehot would expect.
And although I can’t prove it, I do think this is sending distorted signals. If the banking system is under as much strain as the fact that the Fed increased the TAF says, equities shouldn’t be as high as they are.
Eligible TAF collateral is the same as eligible discount window collateral.
And TAF volumes remain small relative to the range of estimates for eventual write-offs.
I feel like we are talking in circles. The discount window was never expected to be used on an ongoing basis, and rates at the discount window (for precisely the Bagehotian reasons discussed earlier) historically were set above Fed fund rates.
The writeoff estimates widely published refer to “banks” but actually lump in investment banks and in some cases also include foreign institiutions which are not eligible to borrow at the Fed,. The losses have been concentrated in investment banks so far and are likely to continue to be.
John Dizard, in a piece I am in the process of putting up, mentions what a mess the credit markets are in in general. Americans seem to see the eruptions here and there, like the auction rate securities market, or the problems with leveraged loans, yet seem in denial about the danger that this represents.
The list of acceptable collateral for TAF and discount window is horrendous. AAA rated CMOs with >10 year duration at 92%. WTF? @80% if there is no market price? No private bank would ever repo that crap, or if they did, the haircut would be tremendous. We already know that the value of these bonds is less than that, so the Fed is loaning more than is secured. In an ordinary repo, that would trigger a margin call, and then failure to deliver on a margin call. The countreparty would dump the collateral to return his funds. In this case, the Fed would lose money. That in itself should tell you something about the health of the banking system.
As regards the journalist, since the Fed deliberately made the system opaque, we do not know who submitted what collateral. However, it’s a pretty safe bet that the banks are using the worst collateral possible. If they have good collateral, especially Treasuries, they can repo them at much more attractive rates with private banks (although I notice that lately the Fed has been repoing significantly below target Fed funds).
As for defaulting, or more accurately, failure to deliver collateral at term, it happens all the time, even with the Fed. Usually, it’s because the back-office screwed up, wrong CUSIP, i.e. of a technical nature. If they catch it early in the day, no problem, but if it’s late there is a default. The guilty party gets fined, but only at target Fed Funds (if memory serves), so the incentive to fix it depends on the availability of $.
So technical defaults happen all the time. Has anyone heard of a real default on repo with the Fed?
P.S. The FHLB also accepts crappy collateral for repo and has increased loans to its banks. Somehow never see that in the newspapers.