Employer-based plans typically have lower premiums because they are, by definition, groups of working-age and (in most cases) working people. They are younger and less sick, so insurance costs are lower. (Dependents are typically more expensive.) This is the reason that companies, once they reach a sufficient size, self-insure, because they are carving out their younger and healthier population from the broader population. Plus, there are savings gained from administrative costs as well.

Selection of plans or providers is not the problem, especially with the current popularity of PPO plans.

I have a really hard time believing that somehow companies self-managing healthcare administration is more efficient than a large group consisting of the entirety of an insurers customers let alone vs. nations with universal healthcare or even ones with single payer private healthcare. And that doesn't even address the many small businesses which need to pay for "individual" plans as if there weren't larger groups with which to distribute the risk.

Even the marketing act of splitting up all these groups up as you describe is a huge waste. Now you have to shop the differences, the companies have to explain the differences, and everyone (insurance co, health provider, patients) gets to track multiple different groups with different rules. It's all a waste compared to fewer larger distributed risk pools - which is the whole point of insurance.