As a rule of thumb, you pay people 25%-33% of what they are worth to you. That other 67%-75% covers the costs of employing them, the overhead of your business--including employees that don't make money for you directly--and some profit for the owners.
If you pay less than 25%, the employee is very likely to find a better deal somewhere else. If you pay more than 33%, you might have occasional difficulty operating your business or giving raises.
There are plenty of potential employees willing to work for less than 25%, and a few more that will demand more than 33%. You should leave them to other employers, that will have other means of monetizing their work.
Candidates are perfectly able to self-sort themselves into a pay range appropriate for them, especially those that know their own worth versus the companies that routinely try to lowball.
This is why one of my stock interview questions is "how do you measure the value to the company of work done by this position?" For me, whether the response is honest or evasive is often more telling than any details given. In my view, the value an employer should bring to the table is the ability to monetize my work to a far greater extent than I could manage on my own. I might be able to provide $2X in value and charge $X for it as a freelancer, so an employer should at least be able to magnify my work into providing $3X or $4X and then pay me more than $X for it. Ideally, the employer squeezes $5X or more of value out of my work, and then pays me $1.6X or more to do it for them.
So what I'm willing to work for is not my lower bound. That bound comes from what I could get working the same job for someone else, or for myself, and that comes from how well different companies can convert my work into cash, and how greedy they are about capturing that additional value for themselves.
I would likely be willing to work for $40/hr. That amount keeps the bills paid. But it won't ever keep me from looking elsewhere for better work.
So my expectation is that employers that comb the globe looking for bargain employees are crap at monetizing the work of their employees. And to some extent, I think that Silicon Valley salaries are way above that 33% threshold, and are dooming those employers to future cash flow problems, unless they actually are able to bring in an extra $1M per developer--which I suppose could really happen with the right unicorn magic. It's really hard to judge, unless you can look at how the company makes money from their developers' work.