I.e., this may be another “tyranny of accounting” problem. See Matt Yglesias for an example.

The usual story with how ending smoking could increase health-care costs is that people live longer, so the medical system has to take care of them when they’re older. (Though without looking at the numbers, I don’t know if this is true. It could be that smokers have to go through long, agonizing cancer treatments that end up costing the same.) By this measure, the best thing for the medical system would be if everyone died in infancy.

Likewise, ObamaCare might make it possible for more people to take part-time jobs now that they don’t need a full-time job to secure insurance; and it might allow people to retire earlier without fear of losing their insurance. I rejoiced almost four years ago that this might happen, and now the CBO thinks that it might.

If we think that people dropping out of the workforce because they can is a bad thing, that is equivalent to saying that it’s bad for people to have more choices. Likewise, if we think that it’s bad for people to live longer and cost society more for their health care, I submit that we’re measuring the wrong things.