Obamacare's Website Gets Its First Performance Metrics. Will They Be Enough?



For the past few weeks, the Obama administration has been promising that that the federal health insurance exchange portal run through HealthCare.gov will work "smoothly" for the "vast majority" of users by the end of the month. But the administration hasn't offered much detail about what that means.

For that, we turn to The Washington Post, which explains that the administration is focused internally on getting 80 percent of users all the way through the enrollment process. In a detail that is suggestive of the managerial failures that led to the disastrous rollout of the exchanges, The Post also reports that the goal is "the first concrete performance standard in the 31/2 years since the government began to design the health exchange." Prior to the launch of the site, the administration had no internal definition of what constituted a working site. And the contractors who built the exchange system apparently did not have clear expectations regarding their work either:

When HHS in 2011 invited contractors to bid on the chance to build HealthCare.gov, the department's "statement of work" did not include requirements typical of many IT contracts in which interested companies must spell out how the system would perform, according to an industry representative close to the project, who was granted anonymity in order to speak frankly. The agreement that CGI Federal, the company chosen as the main contractor, signed on Sept. 30, 2011, also did not contain specific performance criteria, success measures or response times.

Well, the site certainly behaved like one built without any performance criteria.

The revelation of this metric leaves us with several big questions. Is it achievable? And if so, how will we know whether it has been achieved? An 80 percent success rate leaves room for just enough failures that it will be hard to independently verify. And given previous reports by the Post and others that the site is not likely to improve enough to be working by the White House's end-of-month deadline, there's reason to doubt that the goal can actually be reached.

Even if it is achieved, will it be enough? The 80 percent target is simply for users of the exchange to be able to get all the way through the enrollment process. But it doesn't say anything about the accuracy of the enrollment or pricing data that is generated in the process. That could be a big problem, given that the site has been sending bad enrollment data to insurers and has also had trouble with the subsidy calculator that determines premium prices and subsidy levels.

Indeed, if the website becomes generally usable for those who want to enroll but continues to generate bad enrollment and pricing data, we could end up with large numbers of people who think they are enrolled in one plan, at a particular price, but whose enrollment information is never properly transmitted, resulting in mass confusion once new coverage kicks in next year. Hitting this particular goal, in other words, might not really fix anything, and could make things even worse.