For the same reason you do not build a McDonalds in Antarctica. Yes the land is cheap, but you have no market willing to buy your product. The CMC has a mass driver that can deliver the minerals in system, but no method of shipping them to another system. The minerals by themselves are worthless unless you can sell them to a civilian or Imperial market in the system.
Just to be clear, I am not saying I want the requirement for a population of 10 million removed. That seems sensible. What I am saying is that when a system no longer has any CMC sites it should not be part of the list that determines where a CMC is placed. And the home system will always be the first on the list (based on the homeworld population being largest) that means that 50% of the time a CMC will not show up after all homeworld sites have been exhausted. If, as is likely, multiple large population exist in the homesystem then the chance the CMC shows up at all is dramatically reduced.
In my game the list is: Earth, Luna, Mars, Io, Calisto, Faewald, Venus. (Populations over 10 million)
The chance of a CMC is 6% (60,000 income)
The chance of a CMC in Alpha-C (Faewald) is:
6%*50%*50%*50%*50%*50%*50% ~ 0.09% so I'm not holding my breath.
It would be good also to see the CMCs from a worked out site transfered to the existing still working sites. The game says that they will be transfered but I've never actually seen any evidence for it happening. 3 months after one site shuts down, picking numbers out of the air, the remaining sites should gain an even share of the CMCs. So if a complex with 20 CMCs works out a site, and there are 4 remaining working sites in that system they would each gain 5.
Rather humerously...not long after making this post finally my Xenoarch team got their breakthrough. Just after I had made a second team and had forgotten to pick it up. So I can now collect them and bring them back to Earth and then mix and match the teams to get a good one (closer to 100).