Presumably each of the designs has an associated installation and manufacturing cost. Is efficiency here measuring power/cost, power/blade area, power/blade volume, power/land area...?
This does not seem like the kind of principle that would be overlooked by engineers in a technology that is as old as sailing until "just now".
Variable pitch is a well known technique. Flexible blades are used by some types of light aircraft propellers. In the case of the propellers the improvement is not as great as with full variable pitch but the cost/simplicity is better. So the same sort of thing probably applies here. Some technique would have to be found that was cheap/reliable enough to end up with a better result than boring old variable pitch done with bearings and gears.