Effective Fed Funds vs Overnight LIBOR

Here's an article that can articulate a potential response better than I:

http://ap.google.com/article/ALeqM5gXJkHBkXwQWtPp4EaKg_ly_7cM_AD93NSRGG0

It also addresses the fact that the overnight LIBOR has come down by almost 50% between yesterday and today.

Let me know what you think.

Thanks for the link.

"The fed funds rate is the overnight rate at which banks lend funds that are held by the Federal Reserve to other banks; the Fed therefore has some control over it. Libor, on the other hand, is the average bank-to-bank lending rate on the wholesale market, and is a better benchmark for global short-term interest rates."

That sentence relates to my question but the author is too vague to answer anything definitively. I understand that the fed's actions serve to bring effective fed funds to the target rate, but it should also influence Libor as a secondary effect. Put differently, the Fed really does have control over Libor, at least in regular markets. My question now is why their control over fed funds remains in the current dislocated market, but their control in Libor has completely evaporated, given that both rates are based on effective unsecured lending?

one of my VP's made me construct this exact graph (before the press) and originally i took a snapshot of sept 30 of each each from 2003 to 2008. the spread was marginal for all of the years and then it rocketed to over 200bps on set 30 2008. he made me update it when there was the 50bps rate cut and the spread widened to closer to 300bps. interestingly enough, the ECB base rate and the UK base rate vs. their libors are only around 100bps as of oct 8, but that might have changed in the last few days. (these are all for 3 month libors btw).

to answer the question, i think it means that the banks don't believe that they can lend out at such low rates because they have their own liquiity concerns. the government can slash rates all it wants, but the risks involved when the banks lend that low is too high. a lot of them would rather hord cash then lend at 1.5%+marginal spread. this is why people are saying the credit markets are drying up. you have to be a sure thing to be able to get money from banks, or anyone for the matter, in this market.

this problem is a lot worse than other problems like the tech bubble and i think the last time this spread drifted from norm is during the asia crisis, but i may be wrong. i am a relatively new analyst, so it'd be interesting for others more experienced members to add some input

Also, don't forget that another reason for the spike is not only that banks are concerned about their own liquidity, but they are worried that the counter-party will face a liquidity squeeze. This is particularly relevant, because it shows that banks are fearful of lending to other banks even on such a short-term basis.

I disagree about the fed having indirect control of LIBOR:

The fed's action to lower the fed funds rate doesn't necessarily have to affect the LIBOR over in the UK. Why should it? LIBOR is derived from the average lending rates of the most credit worthy banks. So although the British Bankers Association gets together every morning with reps from all of the central UK banks, it is ultimately the input of these representatives which is averaged. Fed Funds rate on the other hand is directly regulated by the government (federal reserve). What they say goes. In these times, a government official will have different objectives (providing liquidity), than a representative of a central bank (maintaining a sound deposit base). Fed will be willing to make money cheap (lowering rates), while reps from banks will be interested in making money expensive (discouraging giving it out to other banks for fear of counter-party risk/credit/economic risk).

two.N.twenty:

I am not talking about the fed funds target rate. The fed funds effective rate is the rate at which us depositories lend excess balances to each other, and it is a market rate. But it is indeed targeted by the government. In normal markets, the fed definitely has indirect control over Libor since the Libor-fed funds spread is tiny and stays constant. In this market, however, the spread is huge. Now, my question could also be phrased as, Why has effective fed funds rate remained so low? Why are US depositories willing to lend to each other at such low rates versus the rates that Libor-surveyed banks are? They are all large banks making unsecured loans to one another. Is there something special about the federal reserve system?

RE_Banker:

I agree with your statement "i think it means that the banks don't believe that they can lend out at such low rates because they have their own liquiity concerns."

My question though is more technical. Taking your statement as true, why are the banks in the federal reserve system willing to lend at such low rates? Fed funds has remained low in spite of all the media discussion around unwillingness to one another. These overnight loans, like overnight loans on which LIBOR is based, are unsecured.

the FOMC - hence open market operations, they buy/sell treasuries from the open market to add/remove liquidity from the market, they directly control the fed funds effective rate that's why it sticks pretty close to the target rate.

LIBOR is not controlled because it is a purely market driven inter-bank lending rate. Libor - ois (you are right it is the 3m tenors) spread is looked at since it shows you the risk aversion in the market. Libor shows counter party and liquidity risk while overnight index swaps carry the interest rate risk of the fed funds rate.

Why do banks have to borrow at a huge spread to what they technically can in the US? London branches of US and London banks can't borrow directly in the US and there are probably some sort of restrictions on just borrowing in the US and shifting it to London.

I was wondering about this. On the same topic, how is LIBOR pronounced? Is it "lee-bore" or "lie-bore"?

Mykyta, thanks for the reply.

"Why do banks have to borrow at a huge spread to what they technically can in the US? London branches of US and London banks can't borrow directly in the US and there are probably some sort of restrictions on just borrowing in the US and shifting it to London."

It seems to me like the spread between what US banks lend to each other at and the rate at which banks that report to BBA lend to each other at (surveyed for LIBOR) ought not be drastically different. You may be right about the restrictions on shifting cash from US to UK and vice versa but why would US banks be so confident about lending to one another in this environment? Why wouldn't they share the same concerns as their UK branches and counterparts?

Leo, it is pronounced lie-bore.

can someone explain ois to me?

OIS stands for overnight index swap. The way I think of it is that parties A and B enter into an OIS in which party A pays to party B the effective fed funds rate every night and party B pays to part A the OIS rate (i.e. the agreed-upon set rate, which is fixed) every night. In reality, the two parties simply settle at maturity.

OIS is basically a guess at what the effective fed funds rate will be for a given period. If it is for a short period in which nobody expects the fed to make any rate cuts or raise the rates, then it should be pretty close to the fed funds target rate.

One of my questions related to OIS throughout this thread is why the effective fed funds rate has actually managed to be close to the target fed funds rate throughout the period. We have all been hearing about how banks are terrified of lending to one another, so if the fed pumps money into the system, it should get clogged and the effective rate should become substantially higher than the target rate (Libor has increased in this manner). However, this has not happened and the effective rate remains low even though interbank lending has supposedly grinded to a semi-halt.

US banks can borrow at the primary rate (the discount window) at 1.75% right now directly from the Fed, which was a taboo before and sort of acts as a ceiling to interest rates in the US. The discount window is a main reason GS and MS became bank holding entities. The effective rate is regulated by the FOMC as mykyta said "they buy/sell treasuries from the open market to add/remove liquidity from the market, they directly control the fed funds effective rate that's why it sticks pretty close to the target rate". An example might help to explain it a little more. To keep interest rates low , the Fed repo buys securities (usually treasuries, fed agencies and GSE debt) and pays for them by making a deposit to the required reserves account of the underlying bank - I would imagine that the Fed itslef takes a hit if the other end of the repo does not repay, so there is actually no risk for the banks' required reserves.

Darkxfriend, great thread.

I'm afraid the answer to your question has more to do with mathematics then you may have expected. LIBOR as opposed to Effective Funds is set by a disproportionately small number of banks, namely: BoA, JPMorgan, Citi, MUFG, Barclays, CS, DB, HBOS, HSBC, Lloyds, Rabobank, Royal Bank of Canada, Norinchukin, RBS, UBS and West LB. Furthermore only the interquartile mean is calculated and whilst that may help simulate market conditions in normal times, I can't think of any bank in the above set which would be too willing to part with excess cash in times like these.

In the US, effective funds is a weighted average of all transactions arranged by major brokers between the hundreds of banks which populate the country. More competition between lenders (although, again, less of a valid reason nowadays), semi-government control and a larger sample set mathematically ... you know the rest.

As for your bonus question, I'm going to feebly guess: ... convention? Maybe Jimbo can share a trader's insight.

breaking them down gets confusing, think of the spread as the pulse of banks. When they're nervous about counter party risk and about their own liquidity, it shoots up and vice versa. Fed funds effective will always be close to target because it's actively managed by the fed, announcing a rate cut does lower the rates but it's open market operations (a nice example by BKBG) that actually adjust it. The discount lending window was an antiquated monetary policy tool before sht hit the fan but now that the stigma associated with borrowing through it is gone, it does in effect put a cap on the fed funds rate because as soon as a bank needs to more cash to fulfill it's reserve requirement with the FED it can just go there and get the cash.

Since OIS is a swap rate, fixed rate that a party agrees to pay in return for receiving a floating rate, that is based on Fed Funds as the floating leg, it won't shoot through the roof because of liquidity conditions and risk because it is managed. I imagine that's why libor is spread to it. Spreading 3m libor to o/n libor won't have the same picture since even o/n libor spikes like crazy, look at the beginning of Oct.

That's my understanding of it at least.

Ok, I think I am getting closer to unhinging my ignorance. I understand the mechanics of open market operations. Let's say the Fed inundates the market with cash by buying treasuries. Let's also say that the banks are borrowing from the discount window. The banks are now sitting on more cash than before as a whole. However, no bank in particular is particularly compelled to lend to another bank because the potential gain of earning interest is not enough to offset fear of lending to a flash-bankrupted counterparty. You have all mentioned very eloquently government intervention, but it seems like government intervention by dumping dollars in is not itself sufficient to prompt banks to lend to each other. Yet, banks in the US have been doing so at a relatively low rate. There is overnight risk, just as with Libor.

BKBG, apologies but I am not sure what the very last sentence about counterparty risk means in your post. Could someone also confirm that borrowing from discount window does not require collateral? If it does, then it is not same thing as going to other banks for unsecured overnight loans, right?

Thanks to all who have responded.

darkxfriend
Repos are typically conducted to fine-tune the level of banking reserves needed on specific day. I meant that there is no risk when banks are doing repos with the fed because they are backed by collateral(below mkt value) and because they borrow from the fed against their triparty agent's required reserves. The triparty agent gets the collateral and the fed credits their required reserves which are used by the other side of the repo. At maturity, if the borrowing bank does not return the cash plus interest I would assume that the fed still increases the triparty agent's required reserves by the same amount and tries to convert the collateral.

There is no collateral needed when you borrow from other banks at the Fed Funds rate, it was just based on their long term relationship - you scratch my back today... That's why I would assume that banks did not do that too much last couple of weeks. There is collateral when you borrow through repurchase agreements and the discount window. When banks borrow at the effective funds rate (EFR) which tries to target the Fed's Fund Rate (FFR) from the fed - they give relatively good collateral (eligible securities I am not sure what is the actual collateral that they accept but think treasuries,US gov agencies and GSE debt, some MBS, etc.) When you borrow from the discount window the Fed allows for crappier collateral but they charge more than the FFR. Additionally, they monitor all banks (CAMEL ratings) and if they think a loan is too risky they might not lend.

There was obviously a decrease for supply of loans as evidenced by very high LIBOR rates last weeks, but there was also a decrease in demand for loans. Actually, most banks had their reserve requirements met because they simply did not lend anyone or pursue any deals - they turned straight to the discount window if they had to. Having all reserve requirements met, a bank does not need to borrow to meet them anymore, which I think is another reason that EFR has been low. With risk of repeating myself and many others on this thread, I believe that lately, it has not been banks lending to each other that kept the effective funds rate low - it was the fed through repo agreements.

Furthermore, how the repo sets the EFR is: there are banks that want cash - 50b for example. They submit collateral and the fed chooses to accept 20b for example. 20b is not a random number - it is the amount the fed needs to lend so it can match its FFR. So, there would be 5b of banks' collateral that would be accepted to repay 1.7% for the cash, 5b with 1.6%, 5b with 1.4%, and 5b with 1.3%, which averages to 1.5% (FFR). Of course, the EFR is almost never the same as the FFR.

I hope this did not confuse you even further:)