I wonder how far the 'fee per published article' gets to covering bare costs for e.g. the journal Nature.
But then, I can't back this up with numbers as they do not publish specifics...
Per that article: "According to its 2013 financials Elsevier had a higher percentage of profit than Apple, Inc."
There's no need to "distribute by volunteers". We have something called the Internet - put the articles on a website, click, and you're done. Indeed, for the most part publishing on paper is a mistake - what we need is the electrons, not paper. Editors aren't free, but they don't cost the money that Elsevier and others charge. Reviews are done by volunteers, almost without exception, so there's no need to pay publishers for review.
For-profit scientific publishers provided an important service in the 1950s. It's not the 1950s any more, and they don't provide any valuable services any more. All the real services - namely the scientific work - are paid by others, and so those others should reap the rewards.
When "best we can" obviously means "best (for us as) we can (scam the marks for)", you are a parasite.