I am putting up my responses ahead of sending a document to the British regulators, the FSA.
A clear factor in the recent calamities has been the lack of expertise at the very top levels of banks. Reviewing the publicly available lists of board members at many firms, I observe that most of them not only have never taken an active role in trading, analysis or risk management, but that today few would even be accepted as a trainee in a less prestigious organisation than they ran into the ground. In effect we had mediocre cavalry officers in charge of nuclear bombers. Fred Goodwin is the most egregious example of this, but he is far from unique.
If RABs have a cut off anywhere near the top level of management then it will be difficult to engineer a bonus system to attract the right type and calibre of staff to lead firms. There will of course be no shortage of applications for roles that pay so well, but quantity is not our objective here. There is also the complex issue of promotion where a moderately senior executive may find a dramatic decrease in the quality of his earnings upon accepting greater responsibility.
This can of course be trivially fixed, by increasing top level remuneration to compensate, and it is quite possible that this is an unintended consequence of such measures.
There is thus an interesting trade off here.
To be effective, RABs must represent a substantial reduction in wealth for poor risk decisions, and so risk-taking executives will require greater compensation for this risk.
Few people currently support the notion that top executives should be paid more, and pushing up the staff costs of banks is not attractive, unless a very substantial benefit can be shown.
There will inevitably be grey areas on who should be covered by these proposals, and the simplest approach is to apply it to everyone in management.
Also a scheme by which everyone is to some extent being treated equally will tend to produce greater cohesion across complex firms, and in any complex financial entity the most important decisions with respect to risk taking are rarely taken at board level, and may be taken two or three levels lower.