Carrots and Sticks


The response to Clearspending has been overwhelmingly positive. People seem to care about government spending data quality to an extent I never would have anticipated. It’s encouraging, and it makes me think we have a real shot at getting these problems fixed.

But there are some people with a different perspective. One of them is Gunnar Hellekson, who wrote a thoughtful blog post about why he disagrees with our approach. Naturally I don’t plan to write responses to everyone who disagrees with us. But we really like and respect Gunnar, and he raised some important points in his post. To wit:

I love quality data as much as the next person, but this is a perfect example of treating government as a vending machine, and it’s poisonous. In 19 months, citizens have access to more data than they ever had before. In some agencies, it takes an average of 43 months to get a new project off the ground. The fact that the US government is even attempting this is amazing.

As I mentioned last year, this is exactly how to prevent innovation in government. If you want change, you have to tolerate imperfection and risk. If every program manager thinks they’ll end up on the front page of the Washington Post or get dressed down onstage at Gov 2.0, nothing will change.

Now, some of the data is ridiculous, yes. But now we know it’s ridiculous. Before some of this data was public, nobody knew it existed. The government employees who worked with it probably assumed it was valid. One of the main reasons to release the data is to permit public scrutiny and that’s exactly what we got. Ironically, Sunlight’s $1.3 trillion discovery is an example of the process working, not a failure.

I agree completely with Gunnar’s concluding thought: Clearspending is an example of transparency working. And I also agree that there are risks to critiquing imperfect data. This is a concern I’ve heard again and again at open gov events, both from people inside and outside of government: if you ask for perfection, you’ll probably get nothing. Public officials have limited resources to spend on opening their information, and ensuring that human-reported data is of high quality is an immense challenge for any organization. Better to encourage the release of imperfect data and avoid scaring off timid bureaucrats — or so the thinking goes.

I think there’s something to be said for this view. But we need to consider what it actually gets us. Is a data set with a few problems better than no data set at all? Well, sure, probably. But what about a data set where 70% of the totals are flawed, as is the case with the data Clearspending examined? At that point, I’m not so sure. If we put that data online, we’re going to lead people toward incorrect answers. Frankly, I’m worried about what happens when people start asking what concrete things the open government movement has accomplished. We need to make sure that the answer isn’t “accidentally misleading a lot of people.”

Now, the response to this is that even if the data’s bad, at least we’ve shifted norms in a positive direction. Open data is good in the abstract, even if this particular data is bad in reality. And this is true: USASpending may not be a very informative website, but it represents a monumental achievement in shifting expectations and attitudes around government data and government itself.

But here I think Gunnar and I simply disagree about the timetable. USASpending has been around for about three years, and in that time the people running it have been able to redesign it twice. I think that was wasted effort: until the data’s fixed, the quality of the presentation is meaningless. It seems as though there’s capacity for improvement available, but it’s not being thoughtfully deployed. That makes me think that this is a fine time for reform. Besides, one should remember that the problems afflicting USASpending predate the site. The data systems that power it have been around since the seventies and eighties, and their quality problems have been known for at least a decade. If it’s too soon to call for fixing them, how much longer should we wait?

For what it’s worth, we spent quite a bit of time speaking with officials at GSA, OMB and various agencies as we worked on this project, trying to better understand the reporting chain. They were uniformly smart, conscientious, amazingly hard-working (you should see the hours these folks keep), and well-informed about the problems we’d found. But when we discussed getting the affected federal spending systems fixed, they started talking in terms of decades. That’s code for “never.” Administrations come and go, and those folks will leave for other jobs before they can see such an effort through.

I don’t think we can accept that. This stuff is important.

Gunnar’s right that we need to focus on praising government when it does a good job — that’s why Ellen singled out NASA and HHS for their fine work during her Gov 2.0 speech. But if OMB feels empowered to redesign USASpending but not to fix it, we need to throw some elbows, shake things up, and see what we can do to get them a mandate for meaningful improvement.

This process can be uncomfortable. It’s not thrilling to us to cause trouble for people who we know are trying hard, and who clearly mean well. But sometimes there aren’t any alternatives. Carrots are nice, but they’re not a substitute for sticks.