Ross Douthat has a useful piece out today on something I’ve been spending a lot of time on lately: what Republicans should do if and when their initial repeal effort fails. Here are some of his policy recommendations:

To address the first problem, Republicans should work to deregulate the new health care exchanges, so that high-deductible, catastrophic coverage can be purchased as easily as comprehensive plans. To address the second, they should propose capping the subsidies for the uninsured, so that they don’t dramatically exceed the value of the existing tax subsidy for employer-provided insurance.

Both of these are constructive ideas. House Republicans can ensure that high-deductible plans are not only eligible for the exchanges, but that the definition of HDHPs is broadened, such that individuals can put more money away in health savings accounts, and with more flexibility around out-of-pocket costs and deductible ceilings.

In addition, capping the dollar amount of subsidies in the exchanges could provide a backstop to the growth of subsidies, in case Republicans are ultimately unsuccessful in repealing PPACA.

Both of these efforts must be pursued with foremost consideration for how they would affect the CBO score of a repeal bill under reconciliation. As a reminder, Republicans will only be able to repeal PPACA via reconciliation if each provision of their repeal measure contributes to deficit reduction.

Ross goes on to endorse a limited enrollment period, something that we have advocated in this space:

The mandate is a harder puzzle, since it works in tandem with the requirement — popular enough to have many Republican supporters — that insurers cease denying coverage to customers with pre-existing conditions. If you repealed the mandate without repealing that requirement, people could simply wait until they were sick to buy insurance, driving everyone’s prices up.

But Republicans could propose dealing with the same problem in a less coercive way. One alternative would establish limited enrollment periods (every two years, for instance) when people with pre-existing conditions could buy into the new exchanges without being denied coverage. Anyone who failed to take advantage wouldn’t be able to get coverage for a pre-existing condition until the next enrollment period arrived. This would reduce the incentive to game the system, without directly penalizing Americans who decline to buy insurance.

Personally, I think two years is too short of an enrollment period: four-to-five years is better, as the German experience demonstrates. But the idea is sound.

UPDATE: Ramesh Ponnuru is less sanguine about the Douthat proposal.