Insurance Generosity Curves

Economic theory doesn't deal well with discontinuities. Nevertheless, that is one thing that characterizes healthcare, according to a wide swath of the economic literature, summarized by Gruber (2008, Journal of Economic Literature). In fact the relationship between generosity of an insurance benefit and the amount paid is proposed to look something like this:
So, basically, 48 million Americans are on the horizontal segment; they don't pay for insurance, and they do not receive any benefits. Then if they pay some threshold per year (say, 2,000 for themselves and 5,500 paid by their employer as a "non-wage benefit") they get minimal coverage that increases in generosity as more is paid in.

So, what might happen if the "public option" were introduced? What if it was lower quality as critics suggest? It might look something like this:

Thus, there would still be a discontinuity in the relationship, but at the bottom could be lifted without necessarily changing the rest. But, you might say, what about people who get seduced by the cheaper option? Here I ask, "what's the big deal?" The fact that they choose it when there are more generous (and more expensive options) proves that they are better off, on average, by revealed preference, and they have money to save up in case they want to pay in for a big procedure that the big mean bureaucrats won't pay for. The curve might look something like this:

If there are no distortions, then maybe the Feds insure 70 million or so, instead of the 48 million that were uninsured (but they also insure another 45-50 million from other existing public programs for a total of about 120 million people) If there are distortions, it might shift the curved segment like this:

Here, the benefits for people paying might decline, but the question is how much, it is unlikely that rich folks won't still have a "Cadillac option" that is as generous as they want it to be. (Another question is who are these people receiving the lowest benefits "on the new curve?" Are they people who had private insurance before, or are they on medicare/aid or other publicly-funded benefits? Are they people who were already at that level on the curve but decide to pay for the private option in spite of an available public option, i.e. to what extent are the "payers" moving horizontally, receiving the same benefits, versus down, receiving less, but still paying?)

One thing that this doesn't suggest, is that assuming inferior care by the government private insurance would go out of business. For that to happen, it's almost as if you would have to admit that the government plan is at least as good as a marjority of the points "on the curve." If they offer higher quality they should be able to keep charging a profitable price and compete on quality. If you think the government would compete the private insurers out of the market entirely, it's hard to make the case that their quality wouldn't be somewhat comparable.