The Next Obamacare Shoe To Drop: 20 States To Be Hit By The Cadillac Tax In 2025
Posted on April 16, 2015
Healthcare, Fiscal, and Tax 4/10/2015 @ 12:00AM
Forget King v. Burwell. In just three short years, one of the ACA's only real cost control mechanisms takes effect â€“ a 40 percent excise tax on high-value insurance coverage, dubbed the Cadillac Tax. Individual plans that cost more than $10,200, and family plans over $27,500 pay the 40 percent tax on benefits over those amounts. There's a strong likelihood that without changes to current benefits, some of the burden will fall on state and local taxpayers. And that's a good thing.
The Cadillac Tax isn't very popular â€“ though it's perhaps the most important reform plank in the ACA. While conservatives hate Obamacare, the Cadillac Tax is a crude version of something they have long called for â€“ a cap on the tax exclusion for employer-provided health insurance. Decried as a tax by liberals just a few years ago the Senate glommed onto the provision in 2009 as one of the few real revenue offsets in the legislation.
And raise revenue it does, bringing in $87 billion from 2018 through 2025, according to the CBO. It is for just this reason that the tax has to stay in place (and calls to remove it are a terrible idea anyway). This has led to some animosity between unions (one of Democrats' key constituencies) and the administration. Indeed, the excise tax poses a unique problem for unions' â€œgold-platedâ€? health care plans, as my Manhattan Institute colleague, Avik Roy, has noted previously. With benefits that are viewed as being considerably more generous than the private sector overall, unions â€“ public sector unions in particular â€“ expect to get hit hard by the tax.
The Cadillac Tax Isn't a Bad Idea
The goal of the tax in the long-run, of course, isn't necessarily to generate revenue. Instead, it's intended to undo the expensive side effects of tax-advantaged, employer-sponsored health benefits. The idea here is simple â€“ with insurance plans no longer any more valuable than cash wages at the â€œgold-plated margin,â€? the tax will act as a de facto cap on employer-sponsored health benefits. For many employers, this will mean resorting to plans with more cost-sharing, which helps keep premiums below the thresholds. In turn, this can put downward pressure on health care spending.
For non-unionized employers, avoiding the tax isn't easy â€“ selling your workers on a cut in benefits requires a top-notch HR department and great communication with employees. But it's possible. Most employers aren't locked in by long-term contract provisions or constitutional amendments that limit their ability to vary health insurance benefits. Moreover, private employers have been increasing deductibles for some time now, and getting results. The approach has been gradual, true, but compare deductibles to where they were almost a decade ago â€“ average deductibles now stand at $1,217 compared to $584 in 2006 (in unionized plans, the average deductible is now $803, while 2006 values aren't reported). The Cadillac Tax will accelerate this trend; but wages are likely to rise as a result.
Unionized employers, especially those in the public sector, will have a much more difficult time avoiding the tax. For both private and public unionized employers, unions aren't likely to back down easily, and will do what they can to avoid a significant reduction in benefits. Falling unionization rates in the private sector make this less of an issue for private employers however. Public employers, on the other hand â€“ states and localities primarily â€“ are still heavily unionized. This means that when contract negotiations happen, getting the benefit reductions necessary to remain below the Cadillac Tax threshold is much more difficult.
A few caveats are in order before moving on:
1. Unions with multiemployer plans (so-called â€œTaft-Hartley Plansâ€?) may have a â€œrelease valveâ€? of sorts. Under these plans, covered workers can take the coverage from one unionized employer to another, making them ideal for temporary or short-term workers. While union leaders have criticized Obamacare for making these plans less attractive for small employers (many of whom won't be subject to the employer mandate, and may find the exchanges a better option), the law allows these plans to treat all coverage as â€œnon-self-onlyâ€? for the purposes of the Cadillac Tax. In plain English, this means that multiemployer plans are only subject to the higher family-plan limit of the Cadillac Tax ($27,500) in 2018, even if the policy in question is for only one person. That makes the Cadillac Tax less ominous for private-sector unions using multi-employer plans.
2. But there are also some assumptions that may need to be re-evaluated. Most policy wonks likely believe that union plans of all stripes (both private and public-sector) are typically more expensive than non-union health plans. This isn't an unreasonable assumption â€“ the core function that unions serve is to bargain for higher total compensation. Unionized workers receive higher wages (about 27 percent higher according to BLS, unadjusted for differences like skill-level and demographics), and since cash wages are more expensive to offer than health benefits, it wouldn't be far-fetched to assume that unionized workers also receive more expensive health benefits. Under these assumptions, union-based health plans would be hit harder and faster than other health plans because of their generosity.A study in Health Affairs raises some doubts about this assumption. Without going too far into the weeds, researchers from the University of Chicago and UCSF evaluated a survey of unionized health plans and compared them with findings from the Kaiser Family Foundation's Annual Survey of Employer Health Benefits. Not surprisingly, these plans had lower cost sharing, were more comprehensive, and required lower premium contributions from members. But when it came to total premiums, there was only a difference of $5 between the cost of employer-sponsored and union-based plans. If this is indeed the case, then private sector unions might have some more leeway.
The point here is that all union plans are not created equal, and it is indeed possible that not all unions will be hit very hard by the Cadillac Tax. For public sector unions, however, the price tag will likely be more significant.
20 States Get Hit by the Tax
How significant? That's unclear. Much of the specifics on state and local health plans is locked away and unavailable to researchers. There are some round-about ways of thinking through this question, however. For instance, we can look to a Pew Charitable Trusts report on State Employee Health Plan spending. The report, based on 2013 plan-year information gathered and analyzed by Milliman, reports a wealth of data on total and per-employee costs, as well as specific plan information including actuarial value and whether a deductible is present.
The data confirms that public sector benefits are, in fact, more comprehensive than those in the private sector. The national averages for single and single plus dependent coverage were $6,840 and $14,796, respectively. By comparison, private sector coverage averages $5,571 and $13,510 for the two categories according to the Medical Expenditure Panel Survey.*
Because Pew reports both average per employee premiums, as well as breakdowns by the type of coverage, the data allows a rough calculation of how many states' average premiums will be subject to the Cadillac Tax.
Assuming relatively slow premium growth** and 2 percent growth in inflation, 20 states will see their average premiums for single coverage hit or exceed the threshold of the tax. In worst shape are Oregon and Alaska, with premiums for single coverage over $1,000 monthly just in 2013. One of the highest-cost northeastern States, New York, won't quite hit the single threshold in 2018, but by 2025 will be $583 over the threshold (implying a tax liability of $233 per policy).
Things aren't as bad on the family front, with only five states â€“ New Hampshire, New Jersey, Rhode Island, Vermont, and Wisconsin â€“ reaching the threshold by 2025. The reason the numbers look so much better here is largely because the family threshold applies to all plans with more than one additional person, and Pew reports premiums similarly. Breaking out â€œtrueâ€? family plans (with four people, for instance) would likely show that many such plans do in fact hit the threshold.
What's important to remember is that these are average premiums. Typically, states offer at least some choice in plans, and with these choices come different premiums and contribution strategies. That's simply to say that variation exists â€“ there will be policies available that may remain below the threshold, while others don't. This means that asking whether a state has reached the threshold isn't the most useful question â€“ instead, we should be more interested in how many employees are in plans that reach the threshold.
Importantly, all of this assumes that benefits are left unchanged. Is this realistic? Maybe, maybe not. But benefits certainly can change. That will involve tradeoffs, however. The biggest tradeoff will be the generosity of plans, in terms of the actuarial value (this is the average share of costs that a plan is expected to pay for the covered population).
It really doesn't get much clearer than this. For the stats nerds, the r-squared is 0.47. In plain English, that means that a change in actuarial value explains 47 percent of the change in the premium.**** This just drives home what should be a very intuitive and self-explanatory point â€“ the more generous the health plan, the more expensive it will be (all else equal!).
The simple way, then, for states to push the cost of their plans below the threshold of the tax is to reduce plan generosity. Given the state of politics and rhetoric, this is, of course, easier said than done. But there are ways of making this transition a bit more palatable.
Generally speaking, it will have to involve making the tradeoff between wages and health care benefits more salient for public sector employees. In New York City, Mayor De Blasio's deal with the city unions emphasizes this tradeoff, making contractual raises contingent on health care savings (though it's turning out that some of these aren't â€œsavingsâ€? in the true sense of the word).
Fortunately, there are many other tools that states can turn to in order to get the savings needed. Tiered network designs, with deductibles/copays for out-of-network services can help drive utilization to lower-cost providers, keeping premiums lower. Implementing deductibles more broadly (a tougher sell) is yet another useful strategy â€“ about 48 percent of state employees are in plans with no deductible. And for those concerned about high deductibles, a meager $500 deductible won't be the end of the world for patients with chronic diseases, but will absolutely cut premiums and unnecessary health care use.
Of course, simply introducing a deductible won't be enough. Indiana has 90 percent of its employees enrolled in a plan with at least a $1,500 deductible; but because of plan generosity (an 88 percent actuarial value) the state still has a relatively high per employee premium of $1,018 per month.
What this all comes down to is simple â€“ public sector unions have received the kinds of health benefits that the private sector has long abandoned (thankfully) because of their cost. The Cadillac Tax is rightfully designed to make plans that take advantage of an unlimited tax deduction prohibitively expensive. Without changes to benefits, state taxpayers and/or public-sector workers will bear the added costs. Changes are necessary, and have been a long-time coming. Unions should see the coming sea-change as an opportunity to right-size benefits, and raise cash-wages for their members.
*To obtain comparable employee plus dependent costs in MEPS, I averaged the cost for employee plus one and family coverage reported. Pew only reports an â€œemployee and dependentsâ€? category that includes all non-self only coverage. MEPS reports that about half of private-sector workers enrolled in coverage were in single-only plans, while 30 percent were in family plans. This implies that about 20 percent were in employee-plus one plans.
**For Cadillac Tax calculations, I used the lower estimate of cost growth of 4.5 percent annually and 2 percent annual inflation, based on Herring and Lentz's Inquiry article estimating the share of firms subject to the tax. I did not apply risk-adjustments based on age or cost as specified in the statute for the tax. The Cadillac Tax threshold at 2025 ends up being $11,831 for individuals, and $31,899 for families.
***Data for Pennsylvania was unavailable; 49 states were covered in the report.
****Analysis of the cost of health plans in the public sector across states means that no controls for industry are necessary. It's likely (though far from a certainty) that outside of state-specific differences in demographics, employees in the public sector across states share more similarities than workers in different industries. This is all to say that plan generosity may explain less variation on average, after controlling for other systemic differences in the covered population â€“ but there are likely to be fewer systemic differences among state employees.