Grades at INSEAD are calculated based on the z-curve, centered on a mean of 2.5 with a standard deviation of 1 (a curve all students will be intimately familiar with by the end of P1). This means that if you get a 3.5, you were 1 standard deviation above the average grade of the class. A failing grade is one that is 3 standard deviations below the mean (i.e. -0.5; yes, we can get negative grades here...)
INSEAD also has a policy of not revealing grades, and strives to make sure no one fails (or at least very, very few people).
So far, so good.
The result, however, is that professors put every effort into crafting impossibly difficult exams, that have very little, if anything, to do with what we actually studied in class. This, they tell us, is for our own good. With these types of exams, the average grade is ridiculously low, and the standard deviation very high (since some students get abysmal results, while others - who used to be Nobel-prize-level experts in advanced statistics or option pricing models - do quite well). This way, no one fails. For example, if the mean is 50, and the standard deviation is 20, even getting a 0 means you pass (since you are only 2.5 standard deviations below the mean).
Not failing is a good thing, don't get me wrong, but the (unintended?) consequence is that a) students are affected by severe depression after exiting an exam and b) (and this one surely is unintended) there is little incentive to study what we did in class. So basically we finish a period needing happy pills and having learned quite a bit less than we should have.
Of course I'm exaggerating, but still, there's a reason why I'm writing this instead of studying POM right now...