In many cases it's scientists who decide if existing funding is used on many small studies or few large ones, and which studies in particular. In that sense it is important to evaluate if maybe having fewer studies with better statistical significance would be better.
> better statistical significance
I can tell from your comment that you understand this already but we should avoid talking about improving statistical significance when what we really mean is reducing the sampling error or the systematic effects---which would of course actually reduce the significance if the hypothesis is incorrect.
>In many cases it's scientists who decide if existing funding is used on many small studies or few large ones, and which studies in particular.
In my experience it works out more like, scientists make their studies as big as funding will allow for while tracking and justifying every single expenditure to your grant providers who can withdraw funding if they're not happy with the way their money's being used.
You won't find any experts to disagree with you on this part.