The overwhelming majority of people who have private insurance are covered by their employers.  That will still be true under any of the proposed reforms.  So it's worth looking at what happens to premiums in the employer-based market after a given plan takes effect.

If the plan is the Senate plan, then according to the CBO, not much changes.  Average premiums might be as much as 3% lower than they otherwise would be.  On the other hand, they might not change at all.

That's actually rather surprising.  We've been hearing a lot over the last few weeks about the transformative power of the excise tax on high cost plans, which is supposed to really incentivize the kind of delivery services reform that could hold down medical cost inflation.  Indeed, the effect is supposed to be so powerful that almost all the revenue estimated from the excise tax actually comes from employers buying cheaper health insurance than they otherwise would and passing those savings onto employees in the form of wages, which the employees then pay taxes on.  By 2019, that excise tax is supposed to be generating $34 billion a year for the treasury

But according to the CBO, while the excise tax will exert downward pressure on large group health insurance premiums, other factors--like the requirement that children be eligible for dependent coverage up to the age of 26--will push them upward.  The result is a maximum savings of 3%, a minimum of 0%.  The CBO notes that because so many people are affected,  even small changes can produce significant revenue. 

That zero does make me sort of wonder what might happen to all that revenue.  The excise tax is basically the entire revenue side (there are a bunch of provisions that also effect revenues, but they roughly balance out; if the excise tax doesn't raise as much as anticipated, there will be no extra revenue to cover new spending, unless something else also changes).  

The logical chain that leads to the excise tax raising this extra revenue indirectly is economically pretty sound, if complicated.  But it also works the other way.  If other employees are getting more expensive coverage, presumably their wages go down, and they pay less in taxes.  If the total average change in employer premiums is 0%, then I would assume that the total revenue raised by the excise tax will be very close to zero.

Of course, it may be that the people who will be affected by the excise tax are wealthier than those who will not.  Indeed, I assume many of them are.  But the differences would have to be huge for the effects of the excise tax to grossly outweigh the loss of taxable income on the other side of the ledger.  And a lot of people getting Cadillac insurance seem to be public employees and union members who have higher wages for plusher benefits; they're not paying a 40% marginal rate on any extra wages they'll get out of this.

Update:  I got a little clarity on this from a source in the know, and the reason it's close to zero is not offsetting upward pressure.  That's because those tables cover the per-person cost, so even though adding dependents raises the per-policy cost, it pushes down the per-person.  (Note:  the excise tax is leveled per-policy).  The explanation for the zero is simply that all of the effects are small, possibly approaching zero, and so zero is included as the upper bound of the range.

As the report says, this market is so large--a trillion or more--that even small percentages can raise significant tax revenue.  But my source confirmed that the closer the eventual number approaches to zero, the less revenue that excise tax will be raising.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.