Robert Camner asked a good, broad question yesterday that I want to respond to:
I’d be even more interested in your analysis of whether a public option that is designed to compete with private insurance plans in a standardized manner is (or is not) good public policy, which I would define as “the greatest good for the greatest number + providing an appropriate safety net for the most vulnerable
This is a damn good question.
Here is how I think about this problem. I first divide the universe into the subsidized, non-subsidized earning over 400% FPL and then the general population of the state.
From here, I first look at people who are eligible to buy individual market insurance but not able to be subsidized. They are no worse off on a simple model as a new choice is being offered. If it is a better choice (better network, better cost sharing, better customer service, better premium… better is a broad term here), they’ll buy the better choice and their welfare improves slightly. If the public option choice is not better, they are no worse off. For this group, a public option makes no one worse off and potentially some folks better off as well as potentially covering some people who otherwise would not have been covered.
This is the simple case. After this, it gets complicated.
The main variables for the subsidized population are the relative price spreads of all non-benchmark plans to the benchmark. If the public option is priced at or above the current benchmark, the analysis looks a lot like the non-subsidized analysis. The choice space slightly expands and potentially the quality of plans being offered by private insurers increases. It won’t significantly increase enrollment nor lower costs but it is a minor welfare improvement.
If the public option is priced below the current benchmark, but above the current least expensive silver plan, it becomes the new benchmark plan. Enrollment probably will decrease as the cheapest silver plans will become more expensive and it is more likely that the cheapest non-silver plans are also more expensive (holding all else equal). This is based on the strong assumption that the marginal enrollee (ie the person flipping a coin to sign up) is extremely price sensitive and making an ACA plan net of subsidies more expensive will lead to them deciding to not get covered. Individuals who have high medical needs may be better off if there is a significant qualitative difference between the public option and the old benchmark plan. They could see a qualitatively better plan at a lower net of subsidy premium but this will not substantially change the enrollment universe.
Now if the public option is priced below the least expensive silver plan, the analysis is dependent on the local circumstances. If the public option is priced at a wider spread than the previous cheapest silver to benchmark spread, enrollment will increase as net of subsidy silver plans are now cheaper. Total enrollment may or may not change depending on the spread of cheapest overall plan relative to benchmark. Individuals who want/need to stay in above their current plans may be slightly worse off.
If the spread of the now cheapest public option to the new benchmark is less than the previous spread, then enrollment will probably decline.
Now from the perspective of a state citizen who is not on the individual market, a public option probably slightly increases the option value of telling my boss to go shove it and walk out. That is a marginal value gain but it is real. Beyond that, it is a wash assuming that the state can administer the program in a way that does not introduce new on-budget costs that compete against other desired public expenditures.