There exists a paradox in social choice theory (game theory applied to elections), the paradox of voting -- this comes out of Anthony Downs around 1950.
We can roughly consider that the utility value of voting is equal to a formula that looks like: E[voting] = pV - c, where c is the cost of voting (time cost, information cost, etc.), V is the value of your preferred candidate winning over other candidates, and p is your probability of being the pivotal voter. By pivotal voter, I mean that if everyone other than you cast their vote first, and then you cast your vote, the question is would your vote have a capability to change the outcome.
The typical view is that c is small but non-zero, V is reasonably large and by definition positive, but p is infinitesimal. And so, no one should bother voting. And yet, people do vote, an off-equilibrium behavior. And although it could be the case that someone's subjective belief in p is much higher than the true p is, this seems an unsatisfactory explanation because even stupid idiots know that their individual vote is unlikely to make the difference.
There are two major lines of rebuttal to the paradox. The first is that individual equilibrium models are not appropriate for assessing these kinds of questions. The second line of rebuttal is that there must exist other terms in the equation. The most promising of these is that the civic religion we have built around voting (social pressure to vote, personal moral imperative to "do one's part in society", the dumb stickers, how good it feels to stand up for what you believe, etc.) creates a "consumption value" of voting. So maybe the equation is E[voting] = pV - c + d where it is assumed that d >= c for many people, who vote.
I mention all of this because based on your post, p = 0, V = Infinity, c = high, and d = 0. If you are looking for a call to action, all you need to do is consider the great many benefits of changing your d term.
Finally, I think it weakens the approachability of your third paragraph argument (why bother, states should do it instead) that it comes after your second argument (environmentalists don't even help the environment) or your first argument (vegans annoy me because I can't eat gluten, why aren't people listening to me).