Demonstrators in support of the Affordable Care Act (ACA). Photographer: Andrew Harrer/Bloomberg

In a piece I recently posted up here, I made four arguments in support of the so-called Cadillac tax, the 40 percent excise tax on health care premium costs above caps set by the Affordable Care Act (the tax is scheduled to kick in by 2018).

First, at least initially, the vast majority of premium costs will be below the caps, meaning they won’t get hit by the tax. Second, by creating a strong incentive to hold down premium costs, the tax is likely to help slow the growth of premium costs going forward. Third, as employers put less compensation into health benefits, they’re likely to put more in wages. Fourth, the tax is expected to raise $90 billion over the next decade, so those who would get rid of it need to come up with a replacement, ideally one with some of the same incentives just noted.

You’ll rarely win friends defending a tax, and I got predictably critical responses. While some were from grumpy partisans who hate Obamacare, taxes, and especially taxes that fund Obamacare, others were sensible and deserve attention. Part of the problem was shorthand on my part and so I’ll elaborate a bit more here. And on point three above–the wage point–I was too dismissive of a legitimate concern.

First, while according to my numbers (extrapolated from the 2014 Medical Expenditure Panel Survey) few dollars are expected to be above the caps (and thus subject to the tax), a significant number of plans could end up being marginally exposed. Let’s say your employer provided a family plan that cost $27,600 in 2018. Since that’s $100 above the family cap, your employer would owe the IRS $40.

But while I expect few dollars to be above the caps, there may well be a significant number of plans above the caps. As I initially wrote: “While a fair number of policy holders—around 5 to 15 percent, depending on assumptions about cost growth—have plans [above the 2018 caps], only a small share of those costs exceed the caps and will be subject to the tax.”

And since the caps are expected to grow more slowly than health costs, this share will grow. By 2030, more than half of employer plans could be above the cap. In my original piece, I noted the much lower 15 percent of costs I predict to be taxed in that year, because I think that’s the relevant number. What matters to plan holders is the extent of their exposure to the tax. But I still should have underscored the point about how many more plans might be affected.

One of the loudest complaints from critics of my earlier post was that while economists uniformly assume that decreases in employer contributions to health coverage will be passed through one-for-one in wages, that’s not how things work in the real world. While I hedged this point a bit, suggesting any such transfer is likely to be incomplete, I didn’t go far enough.

A 2014 White House document states the following about this part of the argument (h/t: PVdW):

“…standard economics implies that, in the long run, reductions in the cost of providing benefits such as health insurance are passed through to workers in the form of higher wages since employers must compete for workers.”

The key word here is “compete.” Sure, in tight labor markets, workers might have the clout to push for full replacement on the wage side, but many critics told me, and I believe them, that this has not been their experience. Moreover, you might think that the tradeoff would be much more likely in union settings, where union negotiators would insist that higher copays and deductibles be offset with higher wages.

But many union folks told me two things: first, it is becoming very common for employers to raise the Caddy tax as a reason for cutting their health benefits, and second, employers are not uniformly agreeing to offset this cost shift with higher wages.

To be sure, these are anecdotes, and the data show that over the long run, the pass through is operative. But none of these studies are particularly recent, and thus may fail to capture the fact that our labor markets have been slack for years now. As my own work his highlighted, such slack has significantly eroded workers’ bargaining power. And, of course, unions are getting whacked from all sides.

Conditional on tighter job markets, I believe the long-run evidence, but that a) implies short term losses for workers and b) as in so many other debates involving wage outcomes, underscores the importance of bargaining power and full employment.

I should have amped up these nuances in my original piece, but all that said, I strongly stand by my argument. The Caddy tax raises needed revenues to support Obamacare and the incentives it creates will save significant costs relative to the currency system, wherein the tax exclusion of employers’ health care spending leads to wasteful spending, which is one reason why U.S. health care is so much more expensive than that of other countries.

But policy analysts in this space need to keep a close eye on how the tax plays out once it is implemented. If cost shifts are too egregious, if wages fail to respond much to such shifts, if, fundamentally, the affordability of quality coverage is significantly diminished, we will need to quickly adjust the tax, perhaps by raising the thresholds more quickly than planned under current law.

I’m sorry I left these nuances out of my original post, but I’m glad they’re in here now.