It stands to reason, and should be unsurprising, that the vast majority of all relevant US technology companies would be complicit, seeing as they're under direct forced compliance.
There is broad electorate support for what the NSA does ('keeping America safe from terrorists' - Joe Public), which is why it has escaped an incredible scandal almost entirely un-altered. It's the same reason the Patriot Act still exists, despite the fact that you're historically more likely to die by choking on a hotdog than at the hands of a terrorist.
A lot of techies seem to look around at their fellow techies, see little support for ubiquitous surveillance and other erosions of civil liberties, and assume the general population feels the same. They then conclude that the government is doing this for nefarious reasons against the wishes of the people, and that the solution is to make government more representative of popular sentiment.
As far as I can tell, government is already representative of popular sentiment here. If we want to change these things, we need to convince people in general that change is a good idea.
Whether or not they're participating willingly or not is another question.
Breaking the model under which they operate is the only real way to ensure that
Maciej Cegłowski's "The Internet With a Human Face" is one of the best examinations of what's wrong, and how to fix it, I've read (though Bruce Schneier and Eben Moglen also do a good job).
That depends on whether there is another country that they could move their companies to, but which does not have similar laws in place :)
Outside the US, they're subject to flat out illegal hacking, offering impunity by virtue of remote access and state-actor status. Though parties whose confidence is violated in that case aren't gagged from reporting this.
These companies could have done more, but I think your standard for coercion is a bit high.