November 10, 2008
Update on FDIC guarantee fees
On Saturday I noted that details of the FDIC guarantees of fed funds implemented on October 14 could introduce a substantial wedge between the fed funds target and the effective fed funds rate. Rebecca Wilder argues that this could not be affecting the current effective fed funds rate due to details of the "opt out" provision. Here I provide some further discussion of this point.
I believe that Rebecca Wilder is correct that I was misinterpreting the FDIC October 16 technical briefing. A substantially clearer statement appears in the implementation of the interim rule in the October 29 Federal Register:
Beginning on November 13, 2008, any eligible entity that has not chosen to opt out of the debt guarantee program will be assessed fees for continued coverage. All eligible debt issued from October 14, 2008 (and still outstanding on November 13, 2008), through June 30, 2009, will be charged an annualized fee equal to 75 basis points multiplied by the amount of debt issued, and calculated for the maturity period of that debt or June 30, 2012, whichever is earlier. The fee charged will take into account that no fees will be charged during the first 30 days of the program.
That seems to state pretty clearly that a bank will not be assessed the 75 basis point fee on overnight fed funds borrowed prior to November 13. Moreover, on November 3 the FDIC extended the deadline for opting out of the program until December 5.
But this brings me back to my original core puzzle. If bank fund managers indeed understand the interim system to be as described, there should be a perfect arbitrage opportunity from borrowing fed funds from the GSEs at a 0.25% rate, holding them as excess reserves on which they are now paid 1.0% by the Fed, and pocketing the difference. Banks trying to take advantage of this should be bidding up the rate for borrowing overnight fed funds above 25 basis points.
If we have among our readers any fund managers who can tell us why you're not bidding up the price of borrowed funds to profit from this right now, I welcome the opportunity to be educated further.
Posted by James Hamilton at November 10, 2008 09:08 AMdigg this | reddit
I don't have an answer, partly because I remain sceptical as before about your explanation. There must be something else going on.
For starters, what is the proportion of total reserves typically held by the GSEs in this new environment? It would have to be substantial in order for the GSEs to have such a measurable effect on the fed effective rate, which is an average, assuming the GSEs have that sort of effect. I don't think it is substantial. I think that whatever is going on is more general than a GSE specific effect, and that there's another explanation. But I don't know what it is.
Posted by: JKH at November 10, 2008 10:24 AM
Deutsche Bank has some theories here:
Posted by: ccm at November 10, 2008 04:06 PM
Bailouts' Unintended Consequences
Within the next 12 months the yields on treasuries will begin to rise. This is because the fools (China etc.) who have been willing to accept low interest rates will be tapped (China has a stimulus package of their own to fund etc.) out.
This will cause all interest rates to rise, which will make the housing market worse. If it gets too much worse then the stability of the financial system will once again be the main concern.
Shorting treasuries seems like a rational investment, but yields may decrease before they increase.
Posted by: Owner Earnings at November 10, 2008 05:03 PM
Come on, James_Hamilton and Menzie_Chinn. You guys are way behind on the news. I'm interested in hearing your ultra-convincing reason why the entire global economy depends on the Fed not telling who it's loaned $2 trillion to after promising transparency.
I don't need to send you a link to the Bloomberg lawsuit story, do it? Since I'm sure you're on top of it.
Posted by: Silas Barta at November 11, 2008 08:45 AM
I originally submitted the following response to your first post in the RGEMonitor comment thread over the weekend rather than here -- my apologies. It is still relevant to your basic question: why banks are not bidding more aggressively for GSE funds:
"Just a quick clarification about FDIC insurance premiums and the fed funds rate. The new 75 basis point insurance premiums won't go into effect until December, so they are not an explanation for the current low level of the effective funds rate. Our discussion (on the Wrightson ICAP site) of the indeterminacy of the funds rate in that future regime was hypothetical, as we still think there is a chance that the FDIC will choose to exclude overnight fed funds from the unsecured debt guarantee program. As for the current environment, the role of the GSEs and international institutions is in fact a sufficient explanation. Banks have no desire to expand their balance sheets, and so demand a large spread on the transaction before they are willing to accommodate GSEs and others who have surplus funds to dispose of. It's a specific example of a general phenomenon: the hurdle rate on arbitrage trades has soared due to balance sheet constraints. That fact can be seen everywhere from the spread between the effective fed funds rate and the target in the overnight market to the negative swap spreads in the 20- to 30-year range that have appeared intermittently of late. Finance models that are based on a "no-arbitrage" assumption will need to be shelved, or at least tweaked, for the duration of the financial crisis."
Posted by: Lou Crandall at November 11, 2008 07:05 PM
Lou Crandall: Thanks for repeating your comment here, since I don't generally look at the comments posted on other sites that reproduce Econbrowser original content. But I really don't understand your answer. Isn't this play completely risk-free money that you just pick up off the table? What exactly is the bad thing that could happen to a bank if it were to earn this easy profit? I do not understand why banks would be unwilling to let risk-free cash flow to them for doing nothing.
Posted by: JDH at November 11, 2008 07:48 PM
I'm still not buying.
Indeed, why should this necessarily have anything to do with arbitrage? If one group of potential fed funds sellers can already earn a minimum level of interest on excess reserves while a second group can't, then the second group to the extent they are long funds will then simply drive down the fed effective rate to the extent they can find willing buyers of funds. And on the demand side, why would those short funds not seek their requirements from the second group at a rate less than required by the first group, if they can find those sellers? The fed effective rate then will be the average of the rate achieved by those sellers who can achieve more than the minimum rate paid on reserves and those sellers who are willing to settle for a reasonable rate between zero and the minimum rate. This depends on intraday distribution of long and short positions, timing of trades, and the relative fed funds reserve positions and participation of the two groups.
This seems to me to be a more basic supply and demand issue, perhaps with two different supply curves operating. As such, it is independent of the insurance issue. And arbitrage isn't a necessary part of the explanation, whatever the timing of insurance implementation.
As I said in an earlier comment, the Fed could resolve this by paying interest on all balances regardless of the source.
Posted by: JKH at November 11, 2008 10:27 PM
The only idea that I can come up with sounds like a conspiracy theory (and therefore, I regard it highly unlikely): that fund managers at the G.S.E.s are not seeking the best rate of return for their cash, but instead voluntarily accepting a return well below the target rate. Since Fannie and Freddie are 79.9% U.S. government-owned now, what are the odds that there is a depository bank out there that is struggling mightily and the government is using its control of the G.S.E.s to give that bank free money?
Posted by: John Barrdear at November 12, 2008 06:25 AM
JKH: The core question is, How much would a bank that is eligible to earn the 1% on excess reserves want to borrow from an institution that is not eligible if the offer rate is 0.25%? If you borrow $1 B, you get to earn $7.5 M. I you borrow $2 B, you get to earn $15 M. If you borrow $x B, you get to earn $7.5x M. If you get to choose x, what value of x do you choose?
Personally, if I get a free $7.5x M, and all I have to do is specify the value of x, then I choose a really, really big value for x. What's your choice for x?
Supply and demand, yes. But demand is the value of x. And I claim there's excess demand if the offer rate is 0.25%
Posted by: JDH at November 12, 2008 06:49 AM
But what about the balance sheet constraint? If a bank tries to earn the differential between 1%-75bps and a market rate below 25bps, it is increasing its working capital. The economical risk is 0 and should we priced this way but if the accounting rules (I am unsure what they are precisely) force you to keep - say 5% of the amount of the loan - in cash, then you'd better not have any trade that could ear you 1/5% * (1%-75bp-FF effective)... this effect reduces further the lower bound. Would be happy to get your thoughts.
Posted by: Bernard at November 12, 2008 07:59 AM
Bernard: That could be it, though like you I'd like to see this worked out numerically. What exactly are these accounting restrictions and how are they enforced?
John Barrdear: So you're suggesting that the GSEs are being deliberately used to funnel a little money to some specific targets? If so, then shouldn't we be hearing from funds managers that they'd like to borrow at 0.25% but they're shut out of this market?
Posted by: JDH at November 12, 2008 08:12 AM
One thing is that the Effective Fed Funds Rate is based solely on brokered transactions; about 25% of the total market a few years back and I'll bet it's less now (but I won't bet much!).
Crandall is claiming that it's balance sheet constraints. There may be some banks who would love to do this, but are close to maxing out their leverage ratio. While borrowing from the GSEs and lending to the Fed will not affect the Tier 1 Capital Ratio or the Total Capital Ratio (since the loan will be 0% risk-weighted) it will affect the leverage ratio.
Even if this is not a regulatory constraint, it may be a capital market constraint if investors are looking at the ratios without doing too much digging to see what they're made of.
Say you gross up the balance sheet with Fed Funds by $X-billion to earn $7.5X-million p.a. If we assume a P/E ratio of 10, this will increase the market capitalization of the bank by $75X-million. What if leverage concerns decrease market capitalization by $100X-million? Or increase your capital market borrowing cost by $125X-million? What if some politician gets on national TV and attacks your firm (and the Fed) for tightening credit standards while parasitically exploiting a loophole at the expense of TAXPAYERS and GSE MORTGAGE-HOLDERS!!! just so greedy spa-going executives can earn a BONUS!!! and decreases your market cap by another $100X-million? Then you're in the soup.
You are implicitly assuming infinite liquidity - an assumption that is usually OK in Fed Funds and Treasuries, but is showing a little strain at this time. Across the Curve highlighted some anomalies in the short Treasury market that would, in normal times, be arbitraged away as fast as electrons pass through silicon.
I have suggested another possibility: it might be simple administrative nonsense. The huge current volumes of reserves and excess reserves might have maxed out credit lines with existing relationships. Until new pathways are created and lines increased, simple frictional effects will keep the market inefficient.
Posted by: James I. Hymas at November 12, 2008 08:25 AM
JDH: Not suggesting, wondering. I'm afraid that I don't know enough of the particulars of inter-bank lending in the US (or anywhere, for that matter), let alone how the data used to compile the effective funds rate is compiled, to answer your question with any certainty.
I had thought (and if I'm wrong here, I hope somebody will correct me) that inter-bank lending was not done in an open market (like the stock market with bid/ask prices published publicly), but through private trades that are then reported privately to the fed. If that is the case, then your hypothetical fund manager wouldn't even know that they'd been shut out until, like you and I, they observed the effective rate published by the NY Fed at the end of the day.
Posted by: John Barrdear at November 12, 2008 08:48 AM
Well, I work as an economist in a large insurance group, and, being also puzzled by this "conumdrum" I had a few discussions with analysts around me. My interpretation of the events is the following:
(i) Dramatic increase in excess bank reserves due to the Fed?s emergency landing programs
(ii) GSE, supranational institutions do not earn interest on excess reserves
(iii) They are net suppliers of cash, but demand is limited (see point (i) above) so that they have to accept a very low rate on their overnight lending (say 0.1-0.3%)
(iv) Bank balance sheet constraints have limited their ability to arbitrage (i.e. borrow GSE funds at 0.1% and earn 1%, the rate of excess reserves) and thus to bid rates up. In addition, we should keep in mind that banks usually turn overly cautious around year-end...
All in all, the latest move by the Fed to raise the interest rate paid on required and excess reserves to the target rate are certainly designed to increase this arbitrage activity, but given point (iv), it may remain inefficient.
Posted by: FA at November 12, 2008 08:48 AM
James Hymas: Thanks for taking this up in your great post (especially the updates). So if the "balance sheet" story is the best we have, aren't we then forced to view the spread between the effective and the target as a symptom of a profoundly dysfunctional operating environment? It is one thing for the system to tolerate no risk taking. It is another thing for the system to prohibit the earning of risk-free profit.
And the blame you're placing on ... who exactly? Who is refusing to let the leverage ratio increase? Bank management? Creditors? Regulators?
Posted by: JDH at November 12, 2008 09:05 AM
FA: Thanks much for sharing this. If you know or find out any more about who enforces these balance sheet constraints, please let us know.
Posted by: JDH at November 12, 2008 09:08 AM
Yes your model is quite clear and logical. There must be inefficiencies in the market that prevent it from clearing to your model price. The propensity to hoard rather than lend is the core inefficiency; after all that was the reason for the Fed's increase in excess reserve setting prior to the payment of interest on reserves. The associated propensity to avoid the credit risk incurred in such arbitrage pass throughs would be another. But my guess is that there must be some other things going on in the detail of the way in which the market clears between the two groups that we simply don't have information on at this point. I think you've posed the right question to any readers who might have hands on working responsibilities in this area. It would be a nice topic for David Altig to post on. (And I've been wondering if there's any connection with the complication of the gross reserve flows that have been created by the Fed's FX swaps, and the way in which those positions show up in reserve distribution.)
Posted by: JKH at November 12, 2008 09:17 AM
John Bardear: The effective fed funds rate is based on brokered transactions. A funds manager puts in a bid with the broker to borrow funds at a given rate. You know whether or not you got those funds. If you offered more than the ask and didn't get it, then you were shut out.
Posted by: JDH at November 12, 2008 09:35 AM
Who is refusing to let the leverage ratio increase? Bank management? Creditors? Regulators?
At the margins, the leverage ratio is enforced by the regulators. The FDIC definition of 'well capitalized' includes a minimum leverage ratio of 5%; 'adequately capitalized' is 4% (there are risk-weighted constraints as well).
I'd say some of the problem is asymmetry of information. Bank Management doesn't know exactly how investors will react to increased leverage ratios - even if the increase is due to risk-free arbitrage - so they don't take the risk. Why risk annoying the capital markets for a relatively small gain?
Investors look at the leverage ratio (and the other headline ratios) and not much further, very often. They don't even want to talk about "risk-free arbitrage" very much, because the last time they heard that phrase they lost half their money. A lot of it comes down to trust, trust that will be restored only gradually, no matter how many billions are poured into bank balance sheets.
Posted by: James I. Hymas at November 12, 2008 10:22 AM
Oh! There's no reason why the Deutsche Bank explanation of leakage from the T-Bill market can't be part of the explanation as well.
If you're a GSE and have a requirement to keep billions in instant-liquidity vehicles, your options are limited. Even at the best of times, the size of the position will make the commercial paper option chancy at best.
Now that 4-week bills trade at about 0.25%, there's one reason why fed funds traders with a little balance sheet room to play with won't bid more than 25bp for size: because they don't have to.
Posted by: James I. Hymas at November 12, 2008 03:51 PM
James I. Hymas: I'll add my thanks for your thoughts. With respect to the T-Bill market imposing an upper limit on the rate for over-night cash: Isn't that as though the traders are saying "yes, there's an arbitrage opportunity there, but I don't want it because there's an even better one over here"?
Posted by: John Barrdear at November 13, 2008 02:30 AM
Isn't that as though the traders are saying "yes, there's an arbitrage opportunity there, but I don't want it because there's an even better one over here"?
I don't understand the question. As far as I can tell, the traders only have one arbitrage possibility, the one in the FF market. The bill market serves simply to restrict the options of the cash investors.
Posted by: James I. Hymas at November 13, 2008 05:54 PM
A basic question:
Did the Fed raise the interest rate paid on excess reserves solely to attempt to raise the Fed funds rate back to target (ie: raise the floor?)
Or did they do it because they simply need more money to further expand their balance sheet? As Dr. Hamilton and others have pointed out, $2 trillion apparently isn't enough. (in other words: Is the Fed that desperate for more funds with which to buy assets?)
In the Money and Banking class I teach (Mishkin's text, 2007--already obsolete) we discussed this issue today because Mishkin mentions in our current chapter that the Fed might want to start paying interest on reserves like other countries do (again, it's obsolete).
I unfortunately explained how the currently apparent arbitrage opportunity didn't really exist, unfortunately going by the information in Dr. Hamilton's previous post, since corrected.
So, it gives me some solace to know that PhDs are looking at this situation and asking "why?" and the answers aren't so obvious.
Posted by: JTapp at November 13, 2008 06:54 PM
Hello again (FA speaking)!
@JDH: You're welcome. As to the balance sheet constraints, I am not sure, but I can see that the accounting constraints at my company are so great right now that sometimes, we can't do economically relevant trades or are even forced to do economically stupid ones only because of solvency or accounting issues. I guess there are fewer constraints in banks than in insurance companies, but still...
I haven't have the opportunity to discuss this with our chief economist, but I will do so next week. Hopefully, since he is much more experienced than I am, he will have an idea. I may also try to discuss with our asset managers.
@JTapp: well, my guess would be that the Fed has mainly raised the interest rate paid on excess reserves to attempt to raise the Fed funds rate back to target.
Finally, if it may give you more solace, I can tell you that, although I also have a Ph.D in economics (in international trade, economic geography...), I still find the answer not obvious. Moreover, the more I work in the finance industry, the less obvious I find things to be!
Posted by: Frederic A at November 15, 2008 12:02 AM
I have two opinions:
In regards to 1% interest rate and the massive liquidity injection, and if I have my mechanics right, wouldn't the establishment of the 1% interest in a way allow the Fed to subsidize banks in need. The logic goes as follows: banks convert their troubled assets which canít be sold over to the Fed, this creates excess reserves, and these reserves earn a 1% interest. Thus, without a bank doing anything, it has added a 1% return to whatever assets offered up as collateral in the Fed liquidity injections.
In regards to the effective fund rate, I think it underscores the major problem which plagues our economy and causes the Fed to plead with banks to lend: banks are committed to deleveraging. Surely banks have a large list of guaranteed investment opportunities that exceed a .25% arbitrage, but the problem is the banks arenít interested. Indeed, this seems to parallel the stories of good balance sheet, solid earning companies having their credit lines cease. If banks wonít accept and invest virtually free money, imagine how difficult it is to go to a bank asking for a loan.
Posted by: hello friendly at November 17, 2008 07:06 AM