You have hit on one of the core features of not only the piilukirves but its predecessor the Swedish timmerbila. It's a fairly late development in the style, (a claim that dangles unsatisfactorily without a long and involved time-line of the axes' evolution which I'm neither qualified nor inclined to go into), you see appear at a certain point. Earlier piis which follow the Swedish model lack this feature, having a parabolic form when seen from the perspective you show in the picture, (we can see it clearly in one picture I already posted in the pictures thread, pp. 65 or thereabouts), but they eventually caught up and by the time mass production came in the more common form of the Billnäs variations took over.
One speculative explanation is the narrowness of the toe is used to open up the chip and in the progression the heel peels it away clearing the line of sight to advance the cut. I'm not convinced suggesting as it does that the work proceeds more or less along the grain of the wood when we know that in the Finnish technique is to work across grain to produce the channels perpendicular to the timber's length down the entire height of the completed stacked walls. At the same time the cross grain action doesn't rule out this effect of the taper on opening the cut, you would simply advance the cutting point of the blade while maintaining a stationary position within the channel on the timber.
I have spent the week with the blacksmiths trying to reproduce timmerbila in this model and this very point posed one of the many great difficulties in the forging process. As close as we got, the end results seemed to be more top-heavy because the edges were lacking this very significant taper like that.
