ba736

My suspicion is that it is very challenging to create good questions that test the material, are clearly worded, and have distinct correct answers.  If you haven’t tried to design fair tests, it’s easy to underestimate how much work it is to do, and how easily questions can go the wrong way, particularly if you have to print and distribute tests worldwide every few months (as with L1).  Therefore, I suspect the CFAI reuses questions a fair amount and they want to avoid people reading the wording and and saying “oh, it’s the question with the Treynor model, the answer is ‘correlation’.”
With L3, the AM section is harder to defend, but maybe they just make it a blanket policy.