Al Kags posted a really thoughtful analysis of why focusing on quick hit tech projects in the context of open government should be a priority for those engaged in the open government movement.
The same day I read Al Kags’ piece, I stumbled across another useful post from the always-smart Matt Andrews at Harvard, who has been a fierce skeptic of many of the mainstream “governance” indicators for a number of years. Matt’s main point is:
Governance indicators should assess the degree to which organizations structure their authority relationships to produce outcomes–not just the degree to which they comply with a generic set of best practice processes. [emphasis mine]
This seems like common sense but has been a serious methodological challenge for a number of good governance practitioners (including Global Integrity) during the past decade: can one prove the linkages between good governance best practices and the governance outcomes we all seek (improved healthcare, education, public safety, etc)? In other words, is it at all possible to draw the links between an effective ombudsman’s office in a country and whether kids there are receiving a better education because of that office’s work? To date, the answer has generally been “no.”
Why this should matter to the open government movement (in my mind) is that we will shortly be moving from the whiz-bang phase of tools and apps to the “show me the impact” phase…if we’re not already there yet (I’d welcome thoughts and perspective on that).
As Al Kags writes:
As the Open Government Partnership takes hold in the world, it is important that we who work in projects in the sector (and especially those of us that are focused on opening third world governments) do not allow Open Government notions to become cliche. The only way we do that is by minimizing on the strategising and the creation of plans and pilot projects and focus on driving implementation. It must be done fast, effectively and concisely.
I think this approach might actually be a bit of a trap. By focusing primarily on outputs (projects, apps, data portals, etc) rather than outcomes (improved service delivery and the like) we risk losing momentum in a similar way to how the traditional transparency and accountability movement lost momentum: only after the fact coming to terms with whether we could trace real-life impact and demonstrate the power and efficacy of our values. The traditional anti-corruption and transparency movement came to this realization almost two decades late and is currently scrambling to invest in research and analysis to parse the outcomes from the output. I desperately hope the open government movement doesn’t make the same mistakes.
Al Kags seems to move in that direction in his post, noting:
The danger here is that in focusing on the technology, we could miss the point all together and get lost in the forest of the technology project. The simple mitigation against this (that I can think of right now) is speed so that the focus is on usage by the citizens and building the necessary ecosystems that fosters truly open government. We must ensure that we keep our eye on the right ball and that we define the true success of Open by the engagement and involvement of the citizen.
I would extend that much further and argue that the true success of Open is whether citizens’ lives are better because of our Open interventions. Citizen engagement, while certainly a laudable goal in its own right, is hopefully not the only reason we are engaged in efforts to open government. Improved service delivery outcomes are, in my mind, the golden chalice after which we are chasing. Let’s make that the priority and structure our tactics and strategies around outcomes and not just outputs, before it's too late.
There seems to be this notion that governments can leap to outcomes. There are three significant problems with this viewpoint.
First, valid outcomes are particularly difficult to determine in government at the best of times. That’s because there is no bottom line like in the private sector. It is relatively easier to determine whether outcome measurements are relevant in the private sector. Outcome measurements in the public sector suffer from false positives and false negatives.
Second, outcomes are affected by a network of interrelated factors, many of which are outside the control of the government. There is also incremental returns where open data becomes more valuable as more data is added.
Third, governments need to assess the cost of improving outcomes. This means linking inputs (costs) with outputs to outcomes. So, there is no magic step to measuring outcomes that does not include inputs and outputs.