3 ms·

I cannot say exactly for this one, but with the LIBOR scandal numbers were probably not that high. LIBOR was probably manipulated by 1 basis point (0.01%), giv

I cannot say exactly for this one, but with the LIBOR scandal numbers were probably not that high. LIBOR was probably manipulated by 1 basis point (0.01%), give or take. Even with $1bn notional deal, which is a very large one, the difference would be $25k for 3 month payment. I mean, well, it's still pretty good money, but it's not like they made a couple of trillions and only paid a mere half-a-billion in fine.

delaaxe 5y ago
I can't believe for one second that LIBOR was only manipulated for one bip at a time.
- bidirectional 5y ago
  Well that was the case. Multi-bp LIBOR moves are pretty rare outside of exigent circumstances and the mechanics of doing so would be pretty tricky given that outlier submissions are thrown out. Wilmott had a good article recently on the mechanics of the manipulation [1]. [1]: https://wilmott.com/the-libor-fix/ https://wilmott.com/the-libor-fix/
- cm2187 5y ago
  There was two distinct manipulations. The first one occured before the crisis. Swap traders asking money market traders to round their libor submission "the right way". That number got averaged across multiple banks, and the impact on the final fixing was likely of the order of a basis point or perhaps a fraction (per the regulator's report). 1 basis point may be small, but is a lot of money to an individual swap trader given the sort of position they must have had on libor futures (and that's money to the trader, not necessary to the rest of the bank who may have a net position the other way). The second manipulation occured at the height of the financial crisis and is of a different nature. Because banks contributions to libor fixing were public, investors were infering from the submissions whether a bank was struggling to fund itself. At the height of the post-lehman panic, some banks decided to lower their submission to not look weak. In this case the order of magnitude would be much higher, in the 10s of basis points. You could argue the banks had a financial gain, but not from the direct impact on libor (it wasn't clear what their position was, though banks tend to be typically net receiver of libor), rather from the perception of not being in distress to investors.
lordnacho 5y ago
I traded swaps during the period that this was happening. Keep in mind there's a LIBOR settlement every day. Also, there would be non-linear products on it, too. I'm not sure about 1B being a very large amount either. I've heard of individual traders who swung a dv01 of $20M.
- cm2187 5y ago
  And actually what matters isn't the daily libor fixing, it is the fixings used for the future expiries, 4 times a year. I believe that's the ones that were heavily manipulated (by some faction of a basis point but on huge notionals), because you hedge with a single future exposures over a number of dates. It is rather the impact on every day borrowers that was tiny. First they would have to be unlucky enough to have their libor fixing falling on one of these 4 dates, even if it did, the impact would likely be a rounding error.
  - lordnacho 5y ago
    I'm not so sure. I sat on both futures and swaps desks. Certainly a lot went through the futures on those 4 dates, but the swaps business was OTC, settling every day. I could easily see someone with varying settlements pushing it conveniently their way every day. Each customer would lose a little bit but put it down to noise. Interestingly I also did FX and people got done on that as well. There's this daily fixing that's used for a variety of products, and a big player like a bank would have a daily delta that would be very nice to nudge the right way each day. You could actually see at 1630 each day that something odd was going on, and there'd be a rumor to go along with it on the squawk box.
mcguire 5y ago
How many $1B deals did JPMorgan have?
- AlexTWithBeard 5y ago
  My guess would be that their USD rates book is about $10trn. That being said, pretty much by nature of the deal the book is hedged - precisely to limit the exposure to LIBOR movements.

delaaxe 5y ago

I can't believe for one second that LIBOR was only manipulated for one bip at a time.

bidirectional 5y ago
Well that was the case. Multi-bp LIBOR moves are pretty rare outside of exigent circumstances and the mechanics of doing so would be pretty tricky given that outlier submissions are thrown out. Wilmott had a good article recently on the mechanics of the manipulation [1]. [1]: https://wilmott.com/the-libor-fix/ https://wilmott.com/the-libor-fix/
cm2187 5y ago
There was two distinct manipulations. The first one occured before the crisis. Swap traders asking money market traders to round their libor submission "the right way". That number got averaged across multiple banks, and the impact on the final fixing was likely of the order of a basis point or perhaps a fraction (per the regulator's report). 1 basis point may be small, but is a lot of money to an individual swap trader given the sort of position they must have had on libor futures (and that's money to the trader, not necessary to the rest of the bank who may have a net position the other way). The second manipulation occured at the height of the financial crisis and is of a different nature. Because banks contributions to libor fixing were public, investors were infering from the submissions whether a bank was struggling to fund itself. At the height of the post-lehman panic, some banks decided to lower their submission to not look weak. In this case the order of magnitude would be much higher, in the 10s of basis points. You could argue the banks had a financial gain, but not from the direct impact on libor (it wasn't clear what their position was, though banks tend to be typically net receiver of libor), rather from the perception of not being in distress to investors.

lordnacho 5y ago

I traded swaps during the period that this was happening. Keep in mind there's a LIBOR settlement every day. Also, there would be non-linear products on it, too. I'm not sure about 1B being a very large amount either. I've heard of individual traders who swung a dv01 of $20M.

cm2187 5y ago
And actually what matters isn't the daily libor fixing, it is the fixings used for the future expiries, 4 times a year. I believe that's the ones that were heavily manipulated (by some faction of a basis point but on huge notionals), because you hedge with a single future exposures over a number of dates. It is rather the impact on every day borrowers that was tiny. First they would have to be unlucky enough to have their libor fixing falling on one of these 4 dates, even if it did, the impact would likely be a rounding error.
- lordnacho 5y ago
  I'm not so sure. I sat on both futures and swaps desks. Certainly a lot went through the futures on those 4 dates, but the swaps business was OTC, settling every day. I could easily see someone with varying settlements pushing it conveniently their way every day. Each customer would lose a little bit but put it down to noise. Interestingly I also did FX and people got done on that as well. There's this daily fixing that's used for a variety of products, and a big player like a bank would have a daily delta that would be very nice to nudge the right way each day. You could actually see at 1630 each day that something odd was going on, and there'd be a rumor to go along with it on the squawk box.

mcguire 5y ago

How many $1B deals did JPMorgan have?

AlexTWithBeard 5y ago
My guess would be that their USD rates book is about $10trn. That being said, pretty much by nature of the deal the book is hedged - precisely to limit the exposure to LIBOR movements.