I think you're downplaying the actual costs they have quite a lot, which is a shame, because even without that the share of profits they extract is still disproportionate and unjustifiable.
These people spend 0 on content and their infrastructure is easily rivaled by free alternatives.
He is in no way downplaying anything. The whole thing is ridiculous.
Yeah right, that's why SciHub[1], which does it for free has been nearly impossible to maintain. The main difficulty is the cost of hosting, and not at all the mounting pressure from the litigation from the publishing industry.
Everything else they might do - collating papers into a physical journal, complaining about bibliography formats, etc. is superfluous nonsense that nobody really cares about.
Look at arxiv - the only thing it's missing that journals have is reviews.
Yelp -- How are their costs significant? Web hosting is extremely cheap.
Twitter -- How are their costs significant? Web hosting is extremely cheap.
Is the value of journals over arxiv worth the price? (market says yes, but markets are sometimes wrong).
https://arxiv.org/help/support/faq#2B
https://confluence.cornell.edu/display/arxivpub/arXiv+Budget...
Or SciHub's, for that matter.
But this is one of the biggest sites in the world. Something closer to what Journal Publisher's would need is Arxiv.
Arxiv's budget is $1m of which they spend $60k on hosting and basically all of the rest on staff and staff support.
Eslevier's revenue is £2.5b. billion. Their profit margin is over 50%. What a monumental waste.
That is a pretty significant difference.