I'm not actively going out looking for other direct marketing companies in order to give them my data -- why would I do so with Google?
"This ad is based on emails from your mailbox and information from your Google account. Ads Settings puts you in control of the ads you see."
So clearly they are using your Gmail information for marketing purposes. Are they selling it directly? No. But it seems dishonest to me to claim that they aren't selling it just because there is a middleman (Google) between the marketer and Gmail user.
Google is different in this regard. Since they don't sell raw data, it's easier to opt-out (remove data via the Dashboard[1], delete your account, etc.).
On the one hand your argument makes a lot of sense to me, but on the other hand using an email provider that is not one of the popular ones might in itself make you seem more suspicious.
And intercepting your email correspondence, no matter the provider, might be so trivial to the intelligence agencies at this point that they might be doing so already anyways. It's not unlikely that they'd be 'scanning' Fastmail email more intently than gmail, precisely because of the kinds of people that have security concerns.
This kind of thinking is dangerous and scares me more than the dragnets themselves. There should be nothing suspicious about a citizen exercising their ability to choose from providers. As an analogy, I find it on par with thinking that hippies/minorities/whatever are somehow subversive just because they don't fit your idea of normal.
And yet hippies were monitored, and yet it's plausible that protecting your privacy does make you suspicious.
So I understand that this is scary, but I don't really understand why you consider this kind of thinking dangerous. If it's true, or if there's a good chance that it's true, then shouldn't that knowledge be acted upon, or at least considered, however scary?
Or am I misunderstanding you and are you just stating that in general you hate this situation?
Paying for email service might be a good idea for a lot of reasons, but people have an extremely absurd sense that they're "fighting the man". You know - by paying in US dollars, to a US company, using US banks....
EDIT: Something I did just see is that Fastmail gives you sieve scripting. Since going from hosting my own inbound email to Gmail, I've really missed having a good email sorting language.
We don't have an account with a US bank, though we do have an account in USD, and we're an Australian company, not a US company.
Other than that, our main datacentre is in the USA, and obviously if every US bank decided not to allow transactions to our accounts, we would be in trouble. That would be true regardless of whether we billed in USD or AUD.
The first (and for most users, more important) is about not giving away your data to marketing companies. Switching to Fastmail accomplishes that (probably?).
The second is about securing your data from governments/state actors. That battle seems kind of futile to me. Even if I were running everything on my own server in my own basement, with my own ISP, I'm almost certainly not qualified to sysadmin a machine that needs to be secure against the NSA.