In 'pricing' shield proficiency, I was struck by the fact that a magicuser can get a +1 to dex/str(?) plus proficiency with light armor (which doesnt use a hand and is up to +3 to AC, from memory) for the cost of one feat. So my thought was that shield proficiency should be a considerably lower investment than that. We also discussed it being potentially worth a couple of skills (either in the skilled feat or from a custom written background which is ultimately my preferred option, I think).
My thought regarding the feat comparison would probably be to shift things in the other direction. If a feat that just gives Shield Proficiency seems lackluster (and I agree it does), the solution in my mind would be to add more abilities to the feat, to bring it up to par. Designing a custom Shield feat for non-armored characters would let it be a character designing element.
Whereas trying to create a trade-off to acquire the proficiency via other elements of the system seems dangerous, if only because the same logic could be extended to other feat elements. Why not trade in skills for an armor proficiency? Or for +1 to a stat? Those are also 'partial' elements that come from feats, after all.
All that said, I don't think it will break the system to come up with a way to harvest skills for other bonuses, or create a custom background that goes a little farther afield in the benefits it gives. But it does seem like the sort of thing that can imbalance the system, and would need to be handled with care.