The education parts of the university and the research parts of the university are not staffed separately. In a research university the people who do research are some of the same people who design lesson plans. The idea is that this enhances a student's educational experience.
I agree though, that tuition is indeed too high. But I'm willing to stand by the assertion that our education system is the best in the world too, not just our research system. This is even more true when it comes to our graduate programs.
It would actually be pretty easy to cut salaries without impacting research; pay separately for teaching and research. Teaching a class pays $4-6k/class, or whatever market rate for a PTL is. Grant supported research pays a market rate salary for research.
Then we get teaching for cheap, but don't sacrifice the quality of research.
As for enhancing a student's educational experience, meh. Some people are good at research and teaching. Far more people are good only at one. Because universities want grant money, this usually means good researchers but bad teachers provide teaching (and waste time not doing research). Not sure how this "enhances" a student's educational experience.
Until this happens it's generally not feasible to pay teachers next to nothing and research staff a lot on the same campus. Then all the students just complain they're not being taught by real professors and they truly will pay twice or four times as much at a different school to get taught by research-based professors. (Who often, as you point out, can't lecture for crap half the time.) Universities do some amount of work to ensure that research pays above grade and teaching only faculty get paid on a completely separate grade, but the expectations of the students generally keeps this from happening in a better way.
That said, there are some students that really do benefit from learning with and participating with the top researchers in the world. Those students do need to continue to go to research institutions for their education and do benefit substantially from the research activity on a campus.
Or is the US credentialing system, or perhaps academic research system the best in the world? I don't see a compelling reason why controlling costs on the education side would harm the credentialing and research parts of the system.