ACA Web Site Post-Mortem

Monday, September 21, 2015

Megan McArdle writes an interesting post mortem of the ObamaCare exchange roll-out, in which she notes, among other things, the naivete of the drafters of the ludicrously-named Affordable Care Act:

Why do I say the architects were naive? First of all, because it seems clear that no one -- neither legislators nor administrators -- had any idea that the exchange they wanted was a very hard technical problem. They specified a site that would do real-time verification of identities and subsidies, price search, and handle payments to insurers.

This system had to be built in three years, and moreover, it could not be built the way Silicon Valley would do it: start small, roll something out, see what works and what doesn't, then iterate, experiment, and scale until you finally arrive at the site you wanted to build. No, this site had to work on Day One, in every state in the nation that declined to build its own exchange. That was a very tall order, and no one seems to have given it much thought. Even when a manager in CMS tried to get the administration to scale things back, officials refused, and apparently simply failed to consider the possibility that trying to do too much would mean they ended up with nothing at all. [bold added, link dropped]
Along these lines, C. Northcote Parkinson once noted that it is easier for a committee to approve a nuclear power plant than a bicycle shed. What happened here is comparable, except that this is the government, which can force people to pay for its mistakes and participate in its centrally "planned" program. Note further that the following, which these same people aim to control, are even more complicated than designing a web site: a huge portion of an economy involving a third of a billion, medicine itself, and medical research. It really is lunacy to attempt central planning here.

Unfortunately, conservative critics of the ACA who laughed at the government's inability to get "even" a "website" right missed chances to persuade people at least twice: (1) They lost the moral high ground by failing to note that forcing people to participate in any such program is wrong; and (2) they looked like buffoons when they spent so much time and energy on such a facile version of this criticism. It's no surprise that this roll-out was so difficult, given what McArdle points out, but there was reason to suspect as much all along. After all, the government has successfully performed complex tasks before (e.g., winning two World Wars, and guiding research on the atomic bomb) and daily administers many other things not-badly enough that most people would need convincing just to suspect things could be done better (e.g., roads).

So, many of the critics were themselves naive at best: plainly not knowing why central planning fails, they thought they were "saved by the bell," in the form of the failed roll-out. Now the difficulty of the problem -- on top of any counterexamples anyone might have thought of to the idea that the government is "incompetent" -- makes the idea look plausible that the government can -- with the "right people" -- plan an entire economy. Sadly, I have no trouble believing that a lot of conservatives buy that argument, anyway.

-- CAV

No comments: