MVP

How good would you feel about an IRS website that promised you 90% of your income tax refund if you used it? Or a DMV website that allowed you to have a pretty good shot of renewing your license tabs, but if your birthday was in December you’d find that your purchase couldn’t be completed? Would it be acceptable to use government services that make a “best effort” to follow the law, or would you expect more? Odds are, you’d expect compliance. If you fell in that minority for whom it didn’t work, you probably wouldn’t be okay with that “best effort”. This is the environment that healthcare.gov lives in, and you can’t forget it.

There’s still a lot of controversy about the rollout of Obamacare and the healthcare.gov website, as well there should be. I’d like to go into more depth about a topic I briefly touched upon in my last post: namely, that agile development could have saved healthcare.gov. There are a lot of people arguing that the way it was developed was faulty, and there’s a lot of truth to that. For a particularly good argument, read Clay Shirky. A lot of it boils down to this: development today is incremental and iterative. healthcare.gov should have launched with a vastly pulled-back scope, nailed that, and built from there. Just like lots of other successful web services have done.

In other words, it all boils down to MVP: Minimum Viable Product. If you are a web startup, your minimum viable product may be to send a 140-character message to another person via SMS. That’s it. You launch with that, and add features as you go. Or you launch with “friending” people, and add Farmville later. Be agile, iterate, learn from your fail whales, and move forward. There’s a lot of precedent in the private sector, and government should act more like a business, right? Why couldn’t the government just do that? In fact, didn’t three people just do such a thing? They did it in a long weekend! Stupid lousy wasteful government.

And this is exactly where realities diverge. Health Sherpa, while incredibly clever, is not healthcare.gov and could never be. It doesn’t allow you to buy coverage. It doesn’t allow you to determine which subsidies you are eligible for. It’s a really slick overview that will help the majority of people get a good sense for what is out there, but it doesn’t cover all the edge cases. It wasn’t designed to. It’s a quick and dirty website, perfect for a first iteration, but not good enough for government work, to use the phrase. Because if healthcare.gov worked great at finding subsidies for 90% of the people, but left money on the table for the last 10%, wouldn’t the outrage be just as loud? Wouldn’t people be rightly upset that the website wasn’t working for them as the law intended?

I’ve said many times before (and I’ll say many times again) that Obamacre is needlessly complex, and as a result brought a lot of this upon itself: if there weren’t so many edge cases, it wouldn’t be so hard to design. But I’ve worked in politics, and I know how the sausage factory operates. Policy wonks don’t write laws, legislators do, and they don’t work in a vacuum, crafting finely-worded laws that work flawlessly. I’d argue that the current system really IS the minimum viable product for health care reform: people in the individual marketplace. Going after all healthcare would have been too aggressive, and simply focusing on Medicaid expansion wouldn’t have been aggressive enough. Making the subsidies work as designed for the private health insurance market is as simple as you could get, “simple” being a relative term of course.

Frankly, delaying the rollout, or rolling out something that provided 80% of the functionality, just wouldn’t have cut it. We’d have outrage, just a different form. Agile works great for “good enough”, and maybe one day we will be okay as a country with the government just being “good enough” when it comes to taxes, the military, health care, and so on. I don’t think we are there yet.

Finally, let’s also keep our eye on the prize: improving the health care system in this country. Today’s OECD report is just another reminder of how we are spending BMW money to drive a Kia (no offense to Kia drivers out there). Healthcare.gov is just a step in a probably decades-long revamping of how healthcare works in this country. To give up now would be like giving up on airplanes in 1903 because the Wright brothers barely flew farther than half a football field. This isn’t the end, it’s the beginning of the beginning.