Unfortunately, the only solution I found for this is "charge more", which seems to be psychologically difficult for programmers, with all of patio11's efforts :)
I think "A" players are the best bargain in the software world...
They are, but try explaining to your average PHB who doesn't even begin to understand the complexities of the 'simple' piece of software they intend to write that Charles here can do the job in 3 months and will cost 50K and that the five guys over there in the corner will take the job for 30K and will take a month longer (or so they think). Then, after 30K becomes 90K and the company is slowly sliding downhill that original 3 months and 50K estimate starts to look pretty good, but then the sunk cost fallacy kicks in and for an additional 10K (over the 90K) this time they'll get it right. And then finally the time will be up, the money will be gone and they'll have to shop for someone who will take on the task to fix it on a contingency basis in the next month or so.
This plays out over and over again for some strange reason.
So yes, A players are a bargain, but only after the money to pay for them has been pissed out the window.
1) The market has imperfect information. The expensive developer could be great or terrible, but the same is true for the inexpensive developer. Even if there is theoretically some way to tell the difference, most employers don't have the skills to do so.
2) Bad management is at least as likely to kill a project than bad developers, so even a developer who is 10x in theory will on average deliver much less than 10x in practice.
Which of course isn't true but good luck convincing people otherwise. (Source: regular clean up jobs after all the money has been spent and the company is on the skids and I get called in at a very large multiple of what an 'A' player would have cost to clean up the mess. I don't mind, it keeps me employed and brings home the bacon but it certainly does not seem to be the most efficient way to run an IT department...)