First, labor unions for the most part do not purchase insurance for their members. Unless they are building trade unions the company arranges for the insurance. Building trade unions would need to purchase the coverage when their members don't work for a specific company.
Second, small companies cannot negotiate as if they are major companies.
Third, labor unions do everything they can to get the best wages and benefits for their members. That also means receiving those benefits at the best cost to the company and the members. Local unions utilize resources available from their district/regional and national offices. And companies do shop around to find the best deal that is available.
Fourth, it is not the labor unions or the companies fault that the insurance coverage is all screwed up. But taxing the insurance companies won't solve the problem. If anything, it would probably make it worse.
Fifth, how would implementing an excise tax on employee health care premiums reduce cost benefit the employee? If anything, their level of coverage would be reduce and/or their medical expenses would go up.
Sixth, how does one define what is a Cadillac Plan? What is considered extravagant health benefits?
The study you linked to does not connect the problem with labor unions or the companies.
It’s often assumed that high-cost health insurance plans—sometimes called "Cadillac" plans—provide rich benefits to plan subscribers. Health reform provisions that treat these plans like luxuries may be misguided. Only 3.7 percent of variation in the cost of family coverage can be explained by benefit design (actuarial value). Benefit design plus plan type (HMO, PPO, POS, or high-deductible plans) explains 6.1 percent of this variation. Industry type and medical costs in the region also play a role. Most variation in premiums, however, remains largely unexplained.
From the article "Cadillac Plan Tax Could Backfire, Study Suggest"
The relatively minor differences in deductibles, copays, out-of-pocket maximums and actuarial value of these plans suggests that there might have been major disparities in medical costs from one region to another. But, while large variations in Medicare costs have been exhaustively documented, this study found no positive correlation between the Medicare data and the variation in the cost of family coverage. This may be attributed, the researchers note, to the fact that health plans operate in a competitive market, while Medicare does not. Or providers may charge private plans and employers more or less, depending on how much they receive from Medicare. But no one really knows the reason, they point out.
What this means for the Senate reform debate is that, in the absence of more solid evidence, placing an excise tax on high-cost plans might not achieve the desired policy goals and could harm some people who now have fairly average–albeit high-priced–coverage. To avoid unintended consequences, it might be wise for the Senate to consider another source of funding.
The Problem with Taxing Cadillac Health PlansAs demonstrated in this paper, many factors influence spending, including age, gender, chronic conditions, and local pricing and practice patterns. These factors are largely beyond the control of individual employers and their employees.
This new tax could undermine the affordability of mainstream employer health benefits for those groups already facing the highest health care costs, including women, older Americans and those with chronic conditions.