I would say mostly #1, though it's unglamourous; and a bit of #2 for Instaparser.
Pinterest is burning through enormous amounts of VC money.[0]
Betaworks seems to just be selling its Instapaper product, together with the Instapaper team.
Thus I'm guessing Pinterest needed a crew that could do what the Instapaper crew can do, and was willing to pay a lot more for it than it was worth to Betaworks.
I can totally see where that could be worth it to Pinterest. Talent is in short supply. Their revenue per engineer is probably not a good (nor available) metric but I bet their burn rate per engineer matters, and their "VCs are Impressed By That" per engineer is probably a huge deal for them.
Hopefully that crew is getting proper relocation packages.
(Or, of course, I'm totally wrong about everything. I don't actually use Pinterest, because it's too visitor-hostile.)
Oh and too bad about Instaparser, I thought that was the most interesting thing Instapaper had done yet. I'm sure somebody at Pinterest came up with a number for what it would cost to build that internally, and that number was probably close to the purchase price.
[0]: https://www.crunchbase.com/organization/pinterest