I was intrigued to see Gruber link–in his response to Ben Smith–to his May 2009 analysis of how to measure affordability for a national healthcare reform plan. After all, I’ve been debating with people who love to cite Gruber on affordability for months, and I’ve never seen them cite it. Now there are several reasons they might not want to rely on this paper. It might be that he starts out by arguing that you can still call something “affordable” even if it isn’t affordable for everyone.
In considering affordability for a group, we need to establish a sensible benchmark whereby insurance is considered affordable if “most of” a group can afford it. We can disagree about what “most of” means, but it would be wrong to define “most of” only as “very close to 100%.”
This, of course, accepts as a baseline some continued medical debt (at least) or even bankruptcies in your definition of “affordable.”
Or maybe it’s the fact that Gruber insists that health insurance (not care) be considered as the same kind of necessity as food and shelter.
Second, it implicitly assumes that health care is less important than these other categories; that is, that if individuals have to spend their resources on these other categories, then they should not have to spend resources on health care. It is unclear why health insurance should take a lower position on the priority scale than other necessities.
But the thing I’m most troubled by in this paper is something Gruber neglects to mention: real data from MA on the number of people who forgo necessary medical care because it is not affordable.
In March 2009–two months before Gruber wrote this paper–MA released the first results [PPT] of how that state’s health care reform had improved access. It showed that 21% of the total population–and even 12% of children–forgo necessary medical care because they cannot afford it. Of the 21% forgoing care, most (something like 18 or 19%) have health insurance–but it is health insurance they can’t afford to use. In a paper contemplating what constitutes affordability for a national plan that resembles the MA plan in many ways, Gruber uses national Kaiser/HRET data, rather than the MA data that is much more directly on point.
Now, I might excuse other analysts for ignoring the MA results, except for two things. First, Gruber boasts of his involvement in the MA program as part of his explanation for his qualifications for the HHS contracts.
Throughout this year I have provided technical assistance to the administration and to Congress with my micro-simulation model, as well as based on my experience as a member of the Massachusetts health connector board.
Also, when the facts from MA suit his argument, he uses them, as he did in a November analysis of how much the Senate plan would reduce premiums.
So rather than looking at a real world study showing what happens when a program very similar to the Senate plan goes into effect–which shows that a significant number of people can’t afford to use their health insurance–here’s what Gruber says about how out-of-pocket expenses affect affordability.
A very conservative response would be to say that a plan is only affordable if the premiums plus the maximum out of pocket exposure does not exceed available resources. This is very conservative because while premium payments are certain, out of pocket payments are not, and a sizeable majority of enrollees will not reach the out of pocket limit.
Moreover, there is a strong argument that out of pocket costs should not be incorporated into a discussion of affordability of insurance. After all, individuals face more out of pocket risk without insurance than they do with coverage. Thus, if an individual is very ill and faces large out of pocket costs under an insurance plan, they would have faced at least those same out of pocket costs, and likely more, had they remained uninsured. So it would be wrong to say that those out of pocket costs were responsible for making insurance unaffordable. That is, it is nonsensical to argue that very sick individuals cannot afford insurance because they will have large out of pocket costs under the insurance plan; indeed, the problem is that these individuals cannot afford not to have insurance.
This is analysis that Jonathan Cohn, with data from Gruber, expands upon here.
But it all comes back to that underlying premise. So long as you define “affordable” in such a way that accepts ongoing medical debt for at least some of your sample in your definition of affordable, then this approach–looking at total risk, rather than whether insurance equates to care, makes sense. It transforms the question of whether health care (not health insurance) is affordable into one that measures degrees of indebtedness for using health care.
But then again, that’s what a lot of bill apologists do: consistently oversell what this kind of reform does, by conflating health insurance with health care.
Update: Fixed percentages forgoing care for clarity.