Policy problems are interesting in that their best solutions from a rational standpoint are often limited by what is politically feasible. For instance, if we wanted to insure every American, we could easily do so by spending lots of money. Of course, there are many in this country who would oppose the notion that we should want to insure every American. So, convincing people that universal coverage is desirable is the first political hurdle.
After that first hurdle has been cleared, there are others who would argue that we ought to insure every American, but ought to spend as little money as possible to achieve this end. Clearly, there are constraints–insuring more Americans will generally cost more than insuring fewer Americans, although the specific policy details do introduce a range of variability into these costs. The goal, at this point, becomes to understand the tradeoffs involved in extending coverage and working to cover the most people for the least amount of money. If this sounds to you somewhat like a production possibilities frontier in economics, congratulations. In theory, then, we know that when we plot the number of newly insured persons against the level of government spending, there exists a curved line along which we would like to be located.
In the June issue of Health Affairs, Elizabeth McGlynn, Amado Cordova, Jeffrey Wasserman, and Federico Girosi take a look at the health reform that was passed and how it compares to the range of possibilities that were considered along the way. They were able to do this by employing the RAND microsimulation model and tweaking things like the level of the low-income subsidy, Medicaid eligibility limits, penalty amounts for failing to obtain coverage, etc. All in all, they looked at 2,016 different scenarios. (And you thought all we were concerned with was whether or not to include a public option!)
If you really want to dig into the details, I suggest that you go here to read the paper, but I’ll give you the quick results: It would have been possible–theoretically–to cover even more Americans at a lower cost to the government than the current law manages to do. However, this could only be achieved by greatly increasing the penalty individuals pay for remaining uninsured. At the same time, there were a number of other policy options that would have proven far worse–covering fewer people and costing more. And of course, there were options that would have been less costly and covered fewer people or more costly and covered more people.
Overall, this simulation analysis demonstrates two important things: First, from the standpoint of efficiency–defined as cost per newly insured person–the health reform we got was less than “perfect,” but far better than many other options. Second, along a line that represents the particular level of efficiency that the current reform delivers, it would have been possible to insure fewer or more persons than this reform will achieve. In other words, once the appropriate level of efficiency was determined, the scope of the reform was limited by the compromise process of passing legislation. In other words, political feasibility became the enemy of the perfect, but still allowed us to realize the good.