It becomes worth it with scale. A VPS costs more than $5 but you can host a very large number of repos, a dormant repo takes no resources but disk space to host, its only activity that consumes resources, which is less than $5-$10 worth
Anyway, how are you going to pay engineer salaries and 24/7 support staff on a $5-10/mo product? Not to mention, you'd be hosting someone's critical infrastructure so they will have a meltdown when you inevitably make a mistake. How long are you willing to burn money before you've reached a scale where it's profitable?
It's paradoxical how all the money in tech goes to the most useless things that are a net drain on society (social media, ad tech, etc.) and we're willing to pay less for the most important things than we would for a cup of coffee.
At the prices people are willing to pay for SCM, it really has to be a complement to some other product you're offering or part of an integrated suite. It's most likely a dead end on its own.
I guess you could do it... but why? There are a thousand ways to make more money with fewer headaches. I wouldn't get into this business myself and I wouldn't fund someone else either.
Yes, you can host a repository for less than $10, but that’s not the question, is it? Can you host it cheaply enough that a customer paying $10 is a viable and worthwhile business? Can you do it for $5 all in with infrastructure, staff, and general and administrative costs? You won’t have much for a marketing budget so you’ll have to offer those free accounts with some features for viral network growth. Oh, and you’ll have to offer better uptime than GitHub. Still sound like a good idea?
It's the actions. The compute to run them isn't cheap at scale nor is the amount of auditing you need to do for security and compliance if you want enterprise/government customers.