While sci-hub undoubtedly plays a role here, in my field the arXiv is really what has obviated the necessity of journal subscriptions. Essentially all math papers, in both the preprint and post-review form, can be found on the arXiv. I would also emphasize that, with regard to decisions involving journal subscriptions and access, authors generally have very little power. As a researcher, if you have the grant money, you can pay to make your work open access; alternatively, you can restrict yourself to only open access journals (which in some fields, could come at significant opportunity cost).
arXiv is a very important entity here, correct. It performs a different role though.
arXiv positions itself in a place before the journals in the food chain; anyone can upload a paper there, including authors who submit to journals, but all of these papers are "preliminary", as in, "before journal edits". It's questionable if journal edits actually bring any value, but nonetheless, their status is that they're going to be "likely different in some way" than whatever journals publish.
SciHub allows free access to the "final" pieces of work, even if they're equivalent (or even equal) to the "non-final" ones available on arXiv. That's a big mental difference.
Google Scholar scrapes arXiv and coalesces the results with journal updates, i.e. it will show the entry for a new journal article with a clearly visible [PDF on arXiv] link when this is available. Google Scholar also scrapes institutional repos (at universities etc.).
Scholar isn't a 'sexy' thing to work on at Google. I'm surprised it still exists, frankly. Google products that aren't career-builders tend to languish.
