Ezra Klein has written a thoughtful post on Eugene Steuerle’s suggestion that the Affordable Care Act offers a powerful incentive to employers to drop coverage.

How will the new law work? A worker whose household cash income is $60,000 in 2016 and who gets no health benefits from her employer would receive a subsidy equal to approximately $9,000. Because the firm provides no health insurance, it must pay a $2,200 penalty, leaving a net gain of about $6,800. By contrast, a worker earning equal compensation who receives employer-provided insurance would receive a subsidy around $3,500 from the exclusion of health benefits from his taxable wages, leaving him more than $3,000 worse off than his counterpart whose employer offers no insurance. This pattern holds until compensation reaches about $84,000, at which point the two subsidies are about the same. Workers earning more than $84,000 do better under the current employer-provided system than they will under the new system. Except for the employer penalty for larger firms mentioned above, there are only limited mechanisms to stop employers from dropping coverage and allowing their employees to enter the exchange.

I’m not persuaded by Ezra’s response to Steuerle’s argument.

The exchanges are closed to large employers until at least 2017, at which point states get to choose whether to let larger employers participate. What Steuerle and Reihan are talking about, however, is something very different: It’s employers dropping coverage entirely. And I think this unlikely. Consider this: It’s beneficial for many employers to drop insurance coverage now. There’s no law saying they have to offer it. And it’s wildly expensive. But they keep offering it.

A few things: first, Ezra seems to make Steuerle’s case pretty well.

But in a world with exchanges and subsidies, it’s possible that both employees and employers would be better off if employers dropped insurance coverage, raised wages and let their lower-income employees get subsidized on the exchange. The upside to that world is that the employees are getting better insurance at a lower cost. The downside is that increased use of subsidies makes things more expensive for the taxpayers.

This strikes me as mostly right, though I’m not sure I’d characterize employees earning, say, $83,000 in annual compensation as lower-income employees, and Steuerle observes that the current employer-based system offers more generous subsidies only above the $84,000 mark.

I’d word things a bit differently: Firms compete for employees. If, as Steuerle argues, dropping coverage means a firm can offer higher cash wages and the new hire can seek more generously subsidized insurance coverage via an exchange, this would represent a powerful inducement to drop coverage. With the employer who doesn’t drop coverage, you get lower cash wages and comprehensive insurance coverage. With an employer who does drop coverage, you get higher cash wages and comprehensive insurance coverage, courtesy of future taxpayers.

Given that firms compete, Ezra’s argument for why he thinks firms won’t drop coverage doesn’t hold up very well.

My judgment is that a large move in this direction isn’t very likely. Benefits are sticky. Workers don’t like disruption. And new businesses that are hiring the sort of workers who would ordinarily get health benefits will be competing against businesses offering health benefits, and workers will likely prefer that arrangement.

This is actually pretty politically salient: if it really is true that workers will likely prefer health benefits offered by an employer over higher cash wages, that seems like a case for a more incremental approach. If the exchanges work tolerably well, and I assume they will, I have to say, I find it very hard to believe that mere sentimental attachment will lead employers to choose lower cash wages over higher cash wages. Ezra continues:

But if you do see a move in this direction, the policy responses aren’t too complicated: A mixture of a stronger employer mandate and more tax credits for businesses is likely, though I’d certainly prefer that we embrace the end of the employer-based health-care system and put into place policies that accelerate its decline.

Of course, this path would entail higher costs for the taxpayer, as Ezra acknowledges earlier in the post. One of the central arguments made by ACA opponents is that we were creating a creaky architecture that would cost more than advertised.

As for the post-2017 period, when large employers will be able to join the exchanges, it’s worth thinking through the likely costs if more employers join the exchanges than the CBO anticipated. My sense is that employers would prefer not to operate miniature welfare states if they could avoid it.

Basically, I think that a lot of the pro-ACA arguments were tissue-thin, and now they are unraveling. Ezra, to his credit, is forthright about his preference for a more generous, more centralized taxpayer-financed system. My sense is that the ACA architecture will lead to the more generous, more centralized taxpayer-financed system that he’d like to see. And so I think it was perfectly fair for opponents of the ACA to highlight what they saw as the dangers of such a system.