I personally liked the book, but I wasn't convinced that the incentive schemes drive the sub-optimal behaviour he identifies, nor did I agree with his proposed solution (changing what bonuses are based on). Here's what he has to say on his blog today.
The key is not the size of management’s pay, but the incentives that come from bonuses, options and other additions to basic salaries. Things would be greatly improved if chief executives’ incomes were limited to their basic salaries. Suggesting this would produce such a howl of rage that even putting the idea into the public domain would be a useful ploy. Faced with the threat, it may be possible to introduce a code of best practice in which bonuses were dependent on things other than profits. For example, minimum levels of growth in output and investment could be required before bonuses based on profit achievements would be payable.
What I find most interesting about this is his suggestion that a good outcome would be the elimination of variable pay (which is a tougher position than in his book). It's encouraging to see that a shift in sentiment against variable pay is underway in relation to the area where it is prevalent (boardrooms)