Even a reduction in base pay is still not going to solve the true problem, though, which is a combination of the following: a) A trader's downside risk is limi

Even a reduction in base pay is still not going to solve the true problem, though, which is a combination of the following: a) A trader's downside risk is limited at zero (worst case: trader loses his job because he screwed up, moves on to the next sucke...I mean, bank) b) His upside risk is limited only by the amount of risk he is able to swindle the risk metrics into allowing him to take on c) VaR is a really freaking easy metric to game. It's trivial to make it look like a set of risky positions are a lot less risky than they are when viewed through the distorted lens of VaR, thanks to the magic of financial engineering, which lets you create derivatives with risk profiles that match pretty much any curve you can think of. IMO the only reasonable solution is to stop pretending that VaR is a meaningful metric when applied to "sure thing" bets. If a bet is overwhelmingly likely to make a small amount of money, then you can be damn sure that it's got a tiny chance of losing a large amount of money, and VaR ignores this altogether. Traders engaging in this kind of bet either need to be watched by a person that understands what's going on or subjected to a different risk metric that adds a significant uncertainty factor to deal with the fact that we don't really know whether the risk of global financial meltdown tomorrow is .001% or .01% (fake numbers, obviously). Or (perish the thought!) maybe we should start asking ourselves whether it's really such a great idea to pay people enormous base salaries to gamble while simultaneously giving them a cut of the profits, given the fact that as a profession they really haven't proven themselves able to do anything but break even in the long run, which I could easily achieve at a far lower cost by giving a thousand monkeys "buy" and "sell" buttons to push...