Network News

X My Profile
View More Activity

Why I support the excise tax

R.J. Eskow wants me to write a lot more about the excise tax, and in particular about the claims of its opponents, and wonders why I haven't been. It's probably useful to answer the meta-question as well as the policy question, if only in terms of setting out the premises of this blog: I am not comprehensive. I can't pick every fight, and don't want to try. The idea behind this blog is that I do the best I can to understand these debates, and then I try to guide readers through my thinking. And in the case of the excise tax, the best evidence and theory I've seen is that the tax is a good idea.

The excise tax on high-value health insurance will disproportionately affect union workers, and so there's been a large, union-funded campaign to get rid of it. The fact that unions have a dog in this fight doesn't mean their argument is flawed. But it is a reason to be skeptical. And the evidence they've been using hasn't impressed me. The Communication Workers of America have been particularly aggressive on this, and they have consistently referenced a study in which they "estimated the effect of the excise tax on the plans with the greatest numbers of participants offered by CWA employers in 43 states for which we have data." The analysis found that "in 2013, the first year of the excise tax, 26 of 43 plans covering families would be affected by the tax." But is that a bad thing?

Over the course of this debate, I've repeatedly talked with union leaders and analysts who wanted to convince me of their side. One of the arguments they frequently used was that they had negotiated their contracts knowing that a dollar in health benefits was worth more than a dollar in wages, because it's exempt from taxation. That, of course, is exactly the problem. The current system sets up an incentive for workers to prefer that relatively more of their total compensation comes in the form of health benefits than wages. At a time when we need pressures to control costs, that's an incentive to increase them. As the CWA example shows, a lot of people have benefited from this system up until now, but at some point, we need to call a stop to it.

More interesting is a recent paper from Jon Gabel, Jeremy Pickreign, Roland McDevitt and Thomas Briggs arguing that relatively little of a plan's price tag can be explained solely in terms of the generosity of its benefits. "Most variation in premiums," they conclude, "remains largely unexplained."

Opponents of the excise tax have argued that this is evidence for their side. After all, if the tax isn't simply catching generous plans, then surely it's a bad idea, right? I don't see it that way. For one thing, if the insurance market is really so mysterious and inefficient that we can't confidently say why one plan is pricier than the next, that's good evidence that there's a place for strong cost controls penalizing fast growth in premiums. It is, at least, worth a try.

For another, the most worrying part of the paper is largely obviated by the bill. It notes that "for small and midsize firms, one to five sick people may account for 50 percent of spending, and thus greatly affect the cost of coverage." That's why we have exchanges that prohibit risk discrimination and that adjust payments based on the risk burden carried by insurers. If the excise tax convinces small employers to enter into the exchange so they're no longer vulnerable to a single employee falling ill, I count that as a good thing. It's another way in which the excise tax encourages the system away from an existing inefficiency.

Then there are all the analysts who do support the excise tax. Most all of the leading health economists. The Congressional Budget Office, which considers it one of the most effective cost control mechanisms on the table. There's the analysis from the Joint Committee on Taxation that says relatively few will pay the excise tax, but that wages will rise substantially as employers downshift to less expensive insurance. There's the simple theory of the thing, which is, in its most basic form, a way of amplifying the market advantage of insurers who do manage to hold down costs.

Moreover, most of the opposition to the excise tax that I've seen has centered on the fact that some of the plans its taxes might not be that generous. That's because the excise tax has been sold as a tax on generous plans, which, in aggregate, it is. But my support for the tax is not based on some moral objection to generous insurance policies. All employer-provided health-care benefits should be taxed. That goes for generous plans and stingy plans alike. People lucky enough to have a job that provides benefits, or a union contract that made their job into a good job that provides benefits, are doing relatively well for themselves, and should have to pay taxes on all of their compensation, just as a worker at a job with no benefits and no union contract has to pay taxes on all of her compensation. The employer tax exclusion is essentially massive redistribution from people the unemployed and people without good benefits to the employed who have good benefits.

The first-best way to accomplish that goal would be to end the employer tax exclusion. That was my initial preference, and it still is. But unions and corporations got together to kill that. The excise tax is is a second-best alternative. The politics are better because you're "taxing insurers," even though workers are ultimately paying, and because you're focusing on generous plans. But it's actually less populist than capping or ending the exclusion, as that would subject health-care benefits to the progressive marginal tax rates workers pay now, while the excise tax is one flat rate. It won't be the first time in health-care reform I've had to settle for a policy I don't think optimal, and I imagine it won't be the last.

All that said, my argument is not that the excise tax is without problems, or sure to work. I quite like Henry Aaron's idea to "base the tax on high-cost plans not on each company’s actual cost, but on the cost of each company’s plan as applied to a population of standardized age distribution." And I don't deny that the excise tax might fall flat. But right now, we need to err on the side of trying cost controls, not ruling them out. I have no doubt that if the excise tax proves overly aggressive, it will quickly be declawed.

More broadly, what we're seeing in the excise tax debate is that cost control will have losers, and for any specific policy, those losers will not want to move forward. Nor, as some hope, will those losers be exclusively rich people. That's why cost control is going to be so hard. But failing to begin this difficult and unpleasant work will, eventually, make losers of us all.

By Ezra Klein  |  December 23, 2009; 2:30 PM ET
Categories:  Health Reform  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Life after the first year
Next: Letters to health-care Santa: Free our data!


Two points:

1. there is no guarantee that employers won't just pocket the cost savings coming from selecting cheaper plans. In fact I would guess that employers would take that extra money and not give it to workers.

2. Why not exclude people making less that 100K per year, so working class people are not affected, but the truly gold plated plans are taxed.

I agree that the tax benefits for health care should be removed.

Posted by: srw3 | December 23, 2009 2:37 PM | Report abuse

Excellent column. I don't have a "Cadillac plan," but I do have a reasonable enough plan through my employer. And in all honesty, I don't see why I shouldn't be taxed on it. Not that I would be gleeful about the prospect, of course -- but really: Why shouldn't I be taxed?

Posted by: LynnDee227 | December 23, 2009 2:47 PM | Report abuse

"exactly the problem"

If you truly believe that exempting employer-provided health benefits from taxation is the problem, then the solution should be easy: do away with the exemption and put everybody on the same level playing field.

Of course, pragmatic pessimists can't do such a thing, because it would be political suicide. And of course the liberal solution is to put a bandaid on the problem in the form of an excise tax, complete with justifications and rationalizations to go with.

How about we level the playing field by allowing all health insurance costs go untaxed. That would greatly benefit the low income folks in the 133-400% PL bracket. How unique. Tax breaks for the poor and lower middle class.

But no. We get mandates instead, in the off chance they'll catch some "young invincibles" who would rather spend their money at Patagonia and Starbucks and trips to Cancun than for insurance.

This is the problem with faux "pragmatic pessimist" progressiveness. It sees the problems, ignores the solutions, opting instead for bandaids, because a bandaid is better than growing a backbone and demanding true structural health reform.

Posted by: jc263field | December 23, 2009 2:49 PM | Report abuse

srw3, all available evidence DOES indicate that workers' wages rise more when health care costs don't go up so much. Of course there's "no guarantee" that it will happen again, but just how many things are there guarantees for? There's no guarantee that cap & trade legislation will be a net benefit to the US and the world, but I still believe the evidence I've seen that suggests it will be and hence support the legislation.

Posted by: Chris_O | December 23, 2009 3:06 PM | Report abuse


What evidence? Also the raising wages when employers costs go down is not the same as increasing wages it just changes the balance and adds to taxable income. I know that my former employer didn't raise my salary when I went to a high deductible health plan, even though the HD plan was cheaper for them. The premium cost that I paid went down, but I didn't get any of the benefit that my employer got for me taking a cheaper option. And this was one of the best fairest employers with great benefits: Health, 403b matching, life insurance, dental, etc.

Posted by: srw3 | December 23, 2009 3:21 PM | Report abuse

I support the policy as well.

Where I get hung up is on claiming it about taxing generous plans, which is much less true than taxing individual that get health care in high-cost states.

As I mentioned before, this statement from the other day isn't correct.

"You" probably don't have these plans, which are tilted towards the rich, not the middle class."

I'll give you the benefit of the doubt that it was a momentary lapse falling into the political rhetoric, but that was essentially an incorrect claim.

You can argue, as I would, that its STILL a good thing to tax individuals in high costs states-- it'll hopefully incentivize high cost states to slow their cost growth-- but it unquestionably will be taxing the middle-class in substantial numbers from those states. Which gets to...

"I have no doubt that if the excise tax proves overly aggressive, it will quickly be declawed."

This is the point. The CBO projections that already exist on the excise tax impact are likely "overly aggressive" as you suggest. So when you claim that the legislation is "paid for" when you recognize that elements may not survive in the long-run, we can project how certain states will be disproportionately impacted by this tax, the whole "paid for" but is knowingly politically unstable. We don't need to wait until 2016 before we realize that the excise tax, as structured, isn't likely to hold up. It'll be a step in the right direction, and hopefully down the road the resulting changes will be revenue-neutral, but "paid for" is sold too strong IMO. We're counting on savings from legislation that we know NOW isn't likely to survive at the end of the day.

Posted by: wisewon | December 23, 2009 3:29 PM | Report abuse

"The first-best way to accomplish that goal would be to end the employer tax exclusion. That was my initial preference, and it still is. But unions and corporations got together to kill that."

IMHO, the energy expended by "progressives" to create a public plan would have been better focused on this issue. Mike Enzi's proposal, which would include benefits in taxable compensation accompanied by an above the line tax exemption,would result in no change for those who receive benefits and improve the lot of those who do pay for their own insurance.

The $8500 "cadillac" limit could be used as the maximum exemption and employers could offer plans that would allow their employees to under that or purchase marginally more coverage at their own expense.

That would at least have had the advantage of putting us onto more equitable treatment of insurance expenses and make everyone aware of what is being spent on his/her behalf.

The fact that there would be no material change in the employer paid population and an improvement for those who are paying their own way, should have been a win-win. Instead, we get this misguided excise tax on benefits; which is first and foremost a funding source for subsidies to buy insurance for other people and has very little to do with a coherent health policy.

Posted by: Athena_news | December 23, 2009 3:37 PM | Report abuse

Excellent defense. Thanks Ezra.

Posted by: umesh409 | December 23, 2009 3:38 PM | Report abuse

I have seen the charts correlating health care costs with wages, showing that wages held steady when health care costs rose but wages rose when health care costs were kept under control (by mananged care in the '90s). But I also wonder if under the new social darwinism companies will really raise wages as costs fall? We have to assume that functioning exchanges would make people more apt to change jobs, not stay locked in because of health care, and employers will be competing for at least the most skilled. But if high unemployment persists, will companies behave as they did in the '90s and raise wages? I certainly hope so.

Posted by: Mimikatz | December 23, 2009 3:40 PM | Report abuse


I'm trying to understand to whom the insurance rating rules apply. In the House bill, I thought the 2:1 age rating, no smoker rating applied to everyone. In the Senate, I thought the 3:1 age rating and 1.5:1 smoker rating applied to only individual insurance and the small group market.

If the House rating rules go into effect, wouldn't those in high-risk professions and to a lesser extent, those who are older, receive a substantial cross-subsidy from the rating rules -- enough such that they would still save more even if their plans were hit by the excise tax above some level?

Posted by: BradGabel2002 | December 23, 2009 3:48 PM | Report abuse

Shorter Ezra: almost all representative agent modeling with multiple assumptions unanchored in reality supports the excise tax, therefore I do. And do not look at the man behind the curtain.

Posted by: BruceMcF | December 23, 2009 3:51 PM | Report abuse

"The excise tax on high-value health insurance will disproportionately affect union workers,..."

I would like to see some statistical evidence for this. While it may be true that unionized workers generally have more comprehensive plans (I've seen no quantifiable data) but union contracts have set the pace for employer benefits at much so that 2/3 of employers say that they will be reducing coverage to comply with the excise tax provision.

The most likely outcome of this tax is, whether or not their policies exceed the limit, many employers will use it accelerate the existing trends to reduce coverage, increase co-pays, and/or discontinue contributions to FSAs and HSAs.

This is not saving, it is cost shifting. Employers will spend less and individuals more.

Posted by: Athena_news | December 23, 2009 4:31 PM | Report abuse


Though many academics have come to support the efficiency of the excise tax, there are still fundamental equality problems with it. When you're taxing plans with higher premiums, that translates primarily into a tax on plans with higher medical spending. As Krugman (1998, I believe) points out, 80% of the cost comes from 20% of the beneficiaries. This is even more pronounced in older populations, where the cost of treating chronic diseases can run into the hundreds of thousands of dollars a year on a recurrent basis. When you're taxing expensive plans, you're creating a system where offering little coverage to these individuals is a quick way to reduce lots of cost. While discriminating against high-risk populations is prohibited and closely monitored in Medicare Advantage and Part D, this is not the case for commercial plans in most states. Health reform bills currently being debated do not have a similar regulatory authority nor do they establish a high level of minimum coverage. The risk is that plans will revert to the minimums as they face excise taxes. To the extent that members not facing high-cost conditions do not want to pay substantially higher premiums and taxes for those conditions, they will buy into plans with less coverage. The result is a death spiral for that sort of coverage. Thus, for high cost individuals, the market mechanism exacerbated by the excise tax, could lead to no coverage. Similarly, as we've seen with recent increases in health care costs, plans simply provide less insurance and pass more costs (and risk) directly onto beneficiaries, healthy and sick.

A basic wish-list response would be assessing the excise tax on premiums for a risk-adjusted population, including health status. This way, we're taxing those who provide above average costs to beneficiaries of a particular health status. But to maintain market-wide coverage for high-risk individuals, this doesn't suffice. As soon as some plans begin cutting benefits to high-risk individuals, the average falls and other plans are exposed to greater excise taxes. To do this completely, it would also adjust for the quantity of "core" benefits provided. It would tax plans that provide the same service at higher cost than other plans, rather than just plans that provide greater coverage of core services. This is equivalent to what Medicare Part D does, establishing basic benefits and providing subsidies only for those. For the benefits that are "enhanced", beneficiaries premiums themselves pay the additional premium out of pocket, but they cannot be subsidized with government dollars.

Posted by: GrandArch | December 23, 2009 4:54 PM | Report abuse

You are playing efficiency off justice. Why not move to a system where efficiency and justice are on the same side? Moving healthcare to a system with a reformed payment system and comprehensive national healthcare budget might be that system. Again why not discuss "the best" even if it is not yet on the table.

Posted by: michaelterra | December 23, 2009 5:00 PM | Report abuse

I don't support this idea, in this form.

We've had 30 years of stagnant middle class wages. To the extent that we have managed to get ahead, it has been through maintaining decent health care, not through increased earning power. And now, even though my inflation adjusted economic situation is worse than my dad's was 30 years ago, you effectively want a major tax increase on the middle class.

We've been screwed out of income gains for 30 years and your proposed solution is to complete the screwing process with a massive tax increase to pay for a bill that makes only marginal changes, if any, to the people being taxed. 30 years of being told we can't have higher wages because health costs have gone up, and now you want to *take away* wages because health costs have gone up.

You can't blame people for perceiving that as a bait and switch. Have you bent your little noggin toward thinking about the effect this would have on Democratic electoral prospects? The first thing they do is pass the biggest tax increase in history, entirely borne by the middle class?

Posted by: pj_camp | December 23, 2009 7:14 PM | Report abuse

"The fact that unions have a dog in this fight doesn't mean their argument is flawed. But it is a reason to be skeptical."

Really? The premise here is that unions are inherently negative - suspect.

Now, I cant say I'm surprised that Ezra believes unions are likely flawed by nature.

So is FMLA worthy of suspicion? OSHA? The Wagner Act?

Posted by: dside | December 23, 2009 8:43 PM | Report abuse

Ezra -
1. 23 of 40 plans affected means lots of people will see benefit cuts. You don't think that's a bad thing. But there are very few people who would agree. After all, did we embark of health care reform to cut back benefits? I didn't notice that was the point.

2.You still blithely point to the unsupported JCT/CBO assumption that wages will rise as employers cut back benefits. Wages should indeed rise when health care costs drop... but the excise tax *increases* employers' costs of providing health care, which could drop wages and increase costs.

3. "It's worth a try"? That's a really bad way to set macro policy. There are better ways to control cost that don't punish the middle class. Grand arch also IDs some smart fairness arguments above.

4. Taxing benefits to raise costs to take away people's coverage is the opposite of reform. It's bad policy and bad politics.

5. I'm very proud of CWA for taking the lead on this (I work for CWA and we're finally getting people to take a hard look a this).

6. You didn't rebut the studies that RJ cited... just blandly dismissed all the real world evidence of the consequences as coming from CWA (us self interested union people who did the on the ground work who outside of the campaign itself did the most work to make health care reform possible).

7. Knocking on us for trying to educate the public that the wonk-named "cadillac" tax actually hits the broad middle class is a really low blow.

8. You know probably just as well as we do that tons of non union workers will get hit. But since we don't have data and nobody else cares about this besides us, you will continue to pretend it's just union workers. That continues to marginalize us and allows people to just take shots at a "special interest" without rebutting the substance.

9. What makes you think taking away the tax exclusion will make spending and plans more efficient? Perhaps it will only take away benefits. Employers alreay have a ton of motivation to make plans and spending more efficient... and they haven't been able to do it. Taxing benefits probably won't create much new effieciencies, but it most certainly will take away benefits.

Posted by: PeteSikora | December 24, 2009 12:52 AM | Report abuse

"Why I support the Excise Tax"

Shorter Ezra: "It's a tax, stupid!"

Posted by: WrongfulDeath | December 24, 2009 9:08 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company