Setting the limits is mostly a management issue i'd say.
Just like you can trade your own account and decide not to bet more than so much. Desks directors should decide what risk is acceptable and what is not.
It is then split into traders risk limits
SocGen's risk simulations
- Baltazar
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
Short Oil, Long Vinegar: Salad spread
- polysena
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
"It is then split into traders risk limits " yes that was what I meant.... is there a consideration of the desk directors' expectations about the "quality of the trader?
И ветер, и дождик, и мгла Над холодной пустыней воды.
- aaron
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
There are different levels of limit. In some cases, it's just a question of sizing the trades. Someone has an idea to make money doing something, it gets authorized up to a risk limit, and the trader or team is expected to keep the positions near, but safely below, the limit. Profits are evaluated relative to the limit and, if the ratio is good, you consider increasing the size.
These are internal controls within the desk. If you violate them, your boss yells at you (maybe not, depending on the boss, if you made money). You might get fired, especially if you do it often, or in large size, or lose money (but you can get fired just for losing money; and you can get fired, or yelled at anyway, for being too far below your limits, you get paid to take risk, not watch the screen). You don't go to jail for violating these limits, because your boss is supposed to know enough about your positions not to need to rely on limits. Also, if you want to violate the limits due to market opportunity or unusual situation, you just ask.
Firmwide limits are more formal, and set top-down rather than bottom up. You are not expected to be near the limit, except rarely. They are not explicitly tied to profit, although, of course, the most profitable desks can demand and get the lion's share of the firmwide limit. Going over limits triggers formal reviews and sign-offs which in some cases can go all the way up the line to the firm risk committee, CEO and board. You could be prosecuted for deliberately violating these limits, although most rogue traders actually go to jail for concealing information or lying, with the goal of disguising losses and limit violations.
You can be a rogue trader without violating limits. If you hide losses, you can stay within limits at all times, but still build up a huge total loss. There have been cases like this, but they're less common. As controls improve, it gets harder to hide losses for long enough to do a lot of damage while staying within limits.
Another trick is to game the limits, which may be part of the SocGen situation. Joseph Jett, for example, did trades that exceeded the size of the treasury market (quite a feat) but in such a way that the reporting system netted them down to reasonable size. In fact he claimed (and I believe him) that he was ordered to do the trades in order to reduce the size of Kidder's balance sheet, and thereby hide the risks of the mortgage operation.
These are internal controls within the desk. If you violate them, your boss yells at you (maybe not, depending on the boss, if you made money). You might get fired, especially if you do it often, or in large size, or lose money (but you can get fired just for losing money; and you can get fired, or yelled at anyway, for being too far below your limits, you get paid to take risk, not watch the screen). You don't go to jail for violating these limits, because your boss is supposed to know enough about your positions not to need to rely on limits. Also, if you want to violate the limits due to market opportunity or unusual situation, you just ask.
Firmwide limits are more formal, and set top-down rather than bottom up. You are not expected to be near the limit, except rarely. They are not explicitly tied to profit, although, of course, the most profitable desks can demand and get the lion's share of the firmwide limit. Going over limits triggers formal reviews and sign-offs which in some cases can go all the way up the line to the firm risk committee, CEO and board. You could be prosecuted for deliberately violating these limits, although most rogue traders actually go to jail for concealing information or lying, with the goal of disguising losses and limit violations.
You can be a rogue trader without violating limits. If you hide losses, you can stay within limits at all times, but still build up a huge total loss. There have been cases like this, but they're less common. As controls improve, it gets harder to hide losses for long enough to do a lot of damage while staying within limits.
Another trick is to game the limits, which may be part of the SocGen situation. Joseph Jett, for example, did trades that exceeded the size of the treasury market (quite a feat) but in such a way that the reporting system netted them down to reasonable size. In fact he claimed (and I believe him) that he was ordered to do the trades in order to reduce the size of Kidder's balance sheet, and thereby hide the risks of the mortgage operation.
- Bachelier
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
"But we don't hear about the rogue trader up $10 million either. "
Hmmm. I know a billionaire who is not on anyone’s list (Forbes, etc.), and I know a self-made 100 millionaire who’d rather run naked through the New York Stock Exchange than tell anyone about his good self, how he made his money, etc.
So the rogue trader up $10 mil wants to and has to be observed for us to know, but I expect a lot of them *don’t* want to be observed.
He's on the beach in Rio, and he doesn't talk about this stuff anymore. These cheeky guys are surveying a different field of cheeks within which to take his next plunge.
Also: this is basic newshound: if it bleeds, it leads.
Rogue trader up $10 mil (snnnnzzz, “roundfile”). Rogue trader down $10 mil, same thing.
Rogue trader *up* $5 billion, heck,…five layers above him will be trying to claim credit and back-dating meetings to say it was their idea and that they were in on authorizing it.
To me, it is obvious why we only hear of the big losses.
Very soon $1 bil rogue trader losses will be comfined to the middle section, bottom of the fold, page 5 (that is newspaper jargon for “burying the story”).
Hmmm. I know a billionaire who is not on anyone’s list (Forbes, etc.), and I know a self-made 100 millionaire who’d rather run naked through the New York Stock Exchange than tell anyone about his good self, how he made his money, etc.
So the rogue trader up $10 mil wants to and has to be observed for us to know, but I expect a lot of them *don’t* want to be observed.
He's on the beach in Rio, and he doesn't talk about this stuff anymore. These cheeky guys are surveying a different field of cheeks within which to take his next plunge.
Also: this is basic newshound: if it bleeds, it leads.
Rogue trader up $10 mil (snnnnzzz, “roundfile”). Rogue trader down $10 mil, same thing.
Rogue trader *up* $5 billion, heck,…five layers above him will be trying to claim credit and back-dating meetings to say it was their idea and that they were in on authorizing it.
To me, it is obvious why we only hear of the big losses.
Very soon $1 bil rogue trader losses will be comfined to the middle section, bottom of the fold, page 5 (that is newspaper jargon for “burying the story”).
Okay, if I can turn a sphere inside out with smooth isotopy, how come I can't turn the manifold that is myself inside out to see why my stomach hurts?
- polysena
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
Thank you Aaron for this super explanation... you are helping me more than I can say. very gratefully Polysena.
И ветер, и дождик, и мгла Над холодной пустыней воды.
- pj
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
SocGen's risk simulations
I'd say an example of lucky rogue trader is given in
Ugly Americans
There the author talks about the concept of fuck you money.
Well, they were legal.
Ugly Americans
There the author talks about the concept of fuck you money.
Well, they were legal.
«Да чего там описывать, планировать! Жизнь всё равно богаче». (Саня Радченко about specification writing)