I suspect that the thinking is that if people were to have exercised their options (this is all hypothetical, mind you), they’d have done so relatively evenly throughout the year, rather than everyone having done so at the end of the year.
I’ve always understood this to be the standard approach to computing fully diluted EPS; I don’t know whether it is prescribed by accounting rules or not, but I suspect that it is. Should that be the case, I do not know why another article would advocate using the current (by which I assume you mean year-end) price.
For the exam it hardly matters, for the mechanics will be the same irrespective of the price you use. Just do it according to CFA Institute’s rules: use the average price for repurchases.
What you said does make sense. Though I do wonder, diluted EPS takes into account potentially dilutive securities right? So wouldn’t the focus be prospective, ie. what EPS would be if all potentially dilutive securities were to be exercised right now? In what situations would one be interested in what could have been?
The example I read was in the context of an M&A deal, but I don’t see how that would change things in relation to CFA Institute’s method.
In any case, you’re probably right in that it is the standard approach.
But if everyone exercised their options at the end of the year, then the stock would be around zero days, so the change in the WACSO would be zero. If everyone exercised their options at the beginning of the year, the shares would be around for 365 days, potentially giving the greatest increase to WACSO. (I say “potentially” only because the increase depends on the number of days and on the number of shares, and the latter depends on the stock price; if the beginning-of-year price were very low, then the impact to WACSO could be small despite the number of days.)
In short, taking the average price is something of a compromise. As with all compromises, it’s a little bit fish, and a little bit fowl.