The BBA met on Friday to review how the different "ibor" rates are set. It went something like this:
"Well mates, what do we think?"
"I think it's too bloody late. Why did we have to meet at 5pm on a friday to discuss this piss."
"It's those American gits. They think they're the center of the world, so our BRITISH Banker's Association meeting has to meet when it's convenient for them--noon New York time."
"I thought it was supposed to be noon GMT, I was here five hours ago."
(laughter)
"Yea, it was noon Greenwich Mean Time, by which I mean Greenwich Connecticut."
(laughter)
"Alright boys, settle down. So what do we think. The world's making a big deal about LIBOR rates. Doesn't seem they care much about the other rates."
"So we agree that all the other rates will remain the same? All those in favor?"
"AYE!" (all in unison)
"Well, that's settled."
"On to LIBOR, then. There are several options on the table. 1) we claim to keep a close regulatory eye on the banks submitting rates, which will probably keep libor rates high and not change much. 2) we can add a few American banks to the mix, which will probably lower rates a little bit. 3) we can suggest a market based approach, like that used for Euribor. 4) we do nothing."
"Let the American gits suffer and leave it be."
"I say we don't change anything. It's worked this long, why should we change it now? The system is perfect, we just have to let it shake itself out."
"I say we put more pressure on the banks to submit true libor rates or we switch to a market based structure."
"What, are you limit short eurodollars in your PA or something?"
(laughter)
"uh. . . no. . . I'm not a market participant. . . that'd be. . .uh. . .unethical. . ."
"Why don't we just watch it longer? The panic over libor seems to be settling down. Maybe we can wait it out and not have to do anything?"
"Yea, if we have to change something, we can always tape-bomb them later."
(laughter)
"Preferably we can tape bomb them in the morning so the Americans have to get up in the middle of the night to figure it out. You can imagine Colin Corgan (head US swaps trader, GS) getting that call in the middle of the night and having to throw off some blonde hooker to get to a computer."
(more laughter)
"Settle down now."
"I say it's getting too late. We can let's leave this up to another day and head down to the pub."
"AYE!"
"Okay, okay. Are we all in favor of calling it a day and going down to the pub?"
(it was known to be a rhetorical question so they all file out.)
"Hey, who's going to do the press release?"
"Thanks for volunteering. We'll see you at the pub."
Saturday, May 31, 2008
Monday, May 26, 2008
A Swapper's Paradigm
So a lot has changed since I last blogged. Most notably, the market has found two major deficiencies with the current conventions and ripped a hole straight through them. The first is the mortgage market and how the agencies work to keep order in the mortgage world. I'll write about that another time. The most fascinating change in my opinion (being a rates derivatives guy) is the changes that have happened in libor space.
Libor (the London Interbank Offer Rate), which I've touched on briefly in my Rates "Learn the Lingo" (http://getonthedesk.blogspot.com/2007/06/learn-lingo-rates.html), has seen some dramatic swings. Swap guys pay very close attention to libor because swap payments are set off of libor. While fixed payments are made at whatever rate was agreed upon, the float payments usually reset every three months according to whatever level the three month libor rate set that day.
Libor sets through something of a committee. Basically a bunch of member banks (16, I think) contribute rates at which they say they would offer to lend unsecured cash to other member banks. The top four and bottom four rates are discarded and the simple average is taken of the remaining eight. It is set sometime around 11:45 london time. Many tenors are published from overnight to one year. The most important are usually the overnight rate and the three month rate.
It's worth noting that the rates published by each bank are not necessarily the rates at which they actually lent money nor the rates at which they actually borrowed money (that seems kinda dumb to anyone? Yea, thought it might.). People had pointed out this deficiency before, but it never actually became an issue until recently. Swap and repo guys will know about turn of year funding issues, where year-end (literally dec 31st) and month end funding cost more because companies have to balance their accounting statements. Funding on those days are particularly expensive. The fed has had to flood the financial system with liquidity (read: give away money for free) on these days at times of crises to prevent companies from being unable to fund themselves (thus going bankrupt).
It turns out that this year a whole lot of these sorts of issues were happening and the liquidity of several companies were doubted. We've all heard of Bear Stearns and the "run on the bank" they had, making the older folks remenisce back to the Great Depression days when such runs on banks were common. Those involved also would have seen a similar process (and many rumors) take place with Lehman, although they turned out quite fine.
Libor starting setting higher and higher as banks started to hoard their cash. In fact, it got so bad that at one point the Fed having cut 100bps had completely priced out of the libor market. That is, libor was setting as high as it was when fed funds were 100bps higher. That was both ridiculous and showed how damaged the sytem was. Then the Fed came out with some of their cool toys, most notably the TAF and the dealer lending facility (the name of which evades me at the moment), to inject liquidity without lowering the funds rate. Down came libor again as panic averted and we came back into a "normal" market environment (or as normal as you can say 20bp moves every other day are).
At one point, the allegations came out about banks lying about where they were funding via the libor rates they contributed. Ironically, I think the most publicized version of this allegation came from a Citi analyst. Well, shortly thereafter libor rates started to skyrocket again as scrutiny around libor made member banks carefully set their libor closer to where they actually funded themselves (and citi's new libor rate was much, much higher).
At this point it might be worth noting that or "IBOR" rates aren't as retarded as LIBOR. Some of them, like Euribor, are set via a market rate or market average. That kind of makes sense, doesn't it? To be able to actually see where the rate should be via market transactions. Well, some people really started to notice this as their funding rates went through the roof and trounced their income margins.
Lately a movement has begun to move the derivatives market away from libor. The action itself makes sense, and a gradual transition of swaps and other derivatives away from libor would probably work. Specifically Goldman has suggested that OIS become the new benchmark rate.
What's OIS? OIS stands for Overnight Index Swap. No, that's not helpful. OIS is a swap that goes off of the fed funds effective rate (the rate at which banks lends excess balances to the Fed overnight--basically banks with more cash on hand than required can make a bit of excess interest rate by keeping it at the Fed). The fed funds effective rate is a transactional rate that anyone can observe (well, anyone involved in the markets, at least). The fed funds effective rate usually stays very close to the fed funds target rate, which is the rate the fed announces their policy cuts around at those FOMC meetings.
OIS swaps already trade a bit in the market, but libor swaps still dominate. OIS swaps make sense for people who are hedging, and the uncertanty around reset risk would go away for both dealers and counterparties. The overnight money markets are less active these days and there are usually very few longer dated transactions anyway (for example, why would a bank ever lend for 6months or a year at libor? It seems silly. The only rate really observable was the overnight rate for libor.). OIS, being tied to the specific set of fed funds effective rates, would make the value of interest rate derivatives much easier to predict.
It will be interesting to see if OIS ends up becoming the new libor over the next year or so. It seems the transition will be slow, and it will be a difficult transition to force. If all the dealers agree to trade OIS swaps instead of libor swaps, they'd still have to get customers to convert. For most institutions it probably makes sense to use OIS, especially those who are hedging rate risk for a balance sheet (specifically bank portfolios funding at the fed funds rate, which are a major player in swap space). For other institutions (i.e. industrial companies), it probably doesn't matter as long as they get some sort of interest rate exposure. Personally, I'd say never bet against those Goldman folks.
Libor (the London Interbank Offer Rate), which I've touched on briefly in my Rates "Learn the Lingo" (http://getonthedesk.blogspot.com/2007/06/learn-lingo-rates.html), has seen some dramatic swings. Swap guys pay very close attention to libor because swap payments are set off of libor. While fixed payments are made at whatever rate was agreed upon, the float payments usually reset every three months according to whatever level the three month libor rate set that day.
Libor sets through something of a committee. Basically a bunch of member banks (16, I think) contribute rates at which they say they would offer to lend unsecured cash to other member banks. The top four and bottom four rates are discarded and the simple average is taken of the remaining eight. It is set sometime around 11:45 london time. Many tenors are published from overnight to one year. The most important are usually the overnight rate and the three month rate.
It's worth noting that the rates published by each bank are not necessarily the rates at which they actually lent money nor the rates at which they actually borrowed money (that seems kinda dumb to anyone? Yea, thought it might.). People had pointed out this deficiency before, but it never actually became an issue until recently. Swap and repo guys will know about turn of year funding issues, where year-end (literally dec 31st) and month end funding cost more because companies have to balance their accounting statements. Funding on those days are particularly expensive. The fed has had to flood the financial system with liquidity (read: give away money for free) on these days at times of crises to prevent companies from being unable to fund themselves (thus going bankrupt).
It turns out that this year a whole lot of these sorts of issues were happening and the liquidity of several companies were doubted. We've all heard of Bear Stearns and the "run on the bank" they had, making the older folks remenisce back to the Great Depression days when such runs on banks were common. Those involved also would have seen a similar process (and many rumors) take place with Lehman, although they turned out quite fine.
Libor starting setting higher and higher as banks started to hoard their cash. In fact, it got so bad that at one point the Fed having cut 100bps had completely priced out of the libor market. That is, libor was setting as high as it was when fed funds were 100bps higher. That was both ridiculous and showed how damaged the sytem was. Then the Fed came out with some of their cool toys, most notably the TAF and the dealer lending facility (the name of which evades me at the moment), to inject liquidity without lowering the funds rate. Down came libor again as panic averted and we came back into a "normal" market environment (or as normal as you can say 20bp moves every other day are).
At one point, the allegations came out about banks lying about where they were funding via the libor rates they contributed. Ironically, I think the most publicized version of this allegation came from a Citi analyst. Well, shortly thereafter libor rates started to skyrocket again as scrutiny around libor made member banks carefully set their libor closer to where they actually funded themselves (and citi's new libor rate was much, much higher).
At this point it might be worth noting that or "IBOR" rates aren't as retarded as LIBOR. Some of them, like Euribor, are set via a market rate or market average. That kind of makes sense, doesn't it? To be able to actually see where the rate should be via market transactions. Well, some people really started to notice this as their funding rates went through the roof and trounced their income margins.
Lately a movement has begun to move the derivatives market away from libor. The action itself makes sense, and a gradual transition of swaps and other derivatives away from libor would probably work. Specifically Goldman has suggested that OIS become the new benchmark rate.
What's OIS? OIS stands for Overnight Index Swap. No, that's not helpful. OIS is a swap that goes off of the fed funds effective rate (the rate at which banks lends excess balances to the Fed overnight--basically banks with more cash on hand than required can make a bit of excess interest rate by keeping it at the Fed). The fed funds effective rate is a transactional rate that anyone can observe (well, anyone involved in the markets, at least). The fed funds effective rate usually stays very close to the fed funds target rate, which is the rate the fed announces their policy cuts around at those FOMC meetings.
OIS swaps already trade a bit in the market, but libor swaps still dominate. OIS swaps make sense for people who are hedging, and the uncertanty around reset risk would go away for both dealers and counterparties. The overnight money markets are less active these days and there are usually very few longer dated transactions anyway (for example, why would a bank ever lend for 6months or a year at libor? It seems silly. The only rate really observable was the overnight rate for libor.). OIS, being tied to the specific set of fed funds effective rates, would make the value of interest rate derivatives much easier to predict.
It will be interesting to see if OIS ends up becoming the new libor over the next year or so. It seems the transition will be slow, and it will be a difficult transition to force. If all the dealers agree to trade OIS swaps instead of libor swaps, they'd still have to get customers to convert. For most institutions it probably makes sense to use OIS, especially those who are hedging rate risk for a balance sheet (specifically bank portfolios funding at the fed funds rate, which are a major player in swap space). For other institutions (i.e. industrial companies), it probably doesn't matter as long as they get some sort of interest rate exposure. Personally, I'd say never bet against those Goldman folks.
I'm back
I'd been away from blogging for a while as I spent a lot of time redeveloping our trading desk's risk systems. I think I'm one of the few traders who really get's my hands dirty when new ideas come up. I spent a lot of time rebuilding how we see our risk, how we view the market (i.e. market monitors) and how we build our curves. It was a long process, but I finally feel pretty confortable with our systems. Just needed a bit of a push and support from above to turn everything upside-down.
Unfortunately doing development work is difficult during the trading day, so I spent a lot of my evenings and weekends building tools. It's good to be back in the game though. Trading more actively (and more products) than ever. Hopefully it will give me plenty of material to blog about as well.
No promises that I keep breaking this time, but I will try to remain active in this blog.
Cheers,
QT
Unfortunately doing development work is difficult during the trading day, so I spent a lot of my evenings and weekends building tools. It's good to be back in the game though. Trading more actively (and more products) than ever. Hopefully it will give me plenty of material to blog about as well.
No promises that I keep breaking this time, but I will try to remain active in this blog.
Cheers,
QT
Subscribe to:
Posts (Atom)
