Of course the publishers say they create immense value by typesetting said PDFs, but as technologists, we can clearly see that this is bunk.
Amusingly, when I was a grad student, some friends showed me how they could use LaTeX to produce a manuscript that looked like a spitting image of a Physical Review article.
Well Phys.Rev. makes their LaTeX style file freely available, so it's not surprising. But yeah, it lifts quite an amount of burden from their editors if you already use their style for typesetting.
In maths everyone typesets their own papers. Good publishers let you use their (La)TeX styles beforehand to make sure it will compile finely, after that they do little else.
Web based LaTeX editing, makes it very easy to collaborate. But also they have many standard journal and university templates, and direct submission links to lots of journals to make the whole process of formatting and submitting a paper much simpler.
Yes, this of course raises the question of why some journals charge so much.
Amusingly, when I was a grad student, some friends showed me how they could use LaTeX to produce a manuscript that looked like a spitting image of a Physical Review article.
That was in 1992.