Measuring Governance: What’s the point?

By Alan Hudson — June 3, 2014. 

Over the last 10-15 years, the fact that governance – the institutional arrangements and relationships that shape how effectively things get done – plays a central role in shaping countries’ development trajectories has become widely acknowledged (see for instance the World Bank’s World Development Report of 2011). This acknowledgement has developed hand-in-hand with determined efforts to measure various aspects of governance.

This emphasis on governance and the efforts made to measure its patterns and understand its dynamics is very welcome. There’s no doubt that governance matters and measuring “governance” and its various dimensions can play a useful role in drawing attention to problems and opportunities, in monitoring compliance with standards, in evaluating efforts to support reform, and in informing decisions about what reforms to implement and how.

But in my experience, discussions about governance and its measurement sometimes gloss over a number of key questions (for a similar argument see the early sections of Matt Andrews’ piece on “Governance indicators can make sense”). These include questions about: what is being measured – “governance” is a multi-faceted and rather woolly concept (see Francis Fukuyama’s 2013 piece on “What is Governance?” and various responses); who is going to use the data that is generated; how that data might have an impact; and what results are being sought.

I’ve noticed this most recently in discussions about the inclusion of “governance” in the post-2015 development framework of goals, targets and indicators. From what I’ve seen, the understandable enthusiasm for ensuring that governance gains a place in the post-2015 framework can lead to discussions that: skate over the fact that the evidence that particular forms of governance – often labelled as “Good Governance” – lead to better development outcomes is patchy; fail to effectively grapple with the fact that a cookie-cutter approach to governance is unlikely to work across diverse contexts; pay little attention to the ways in which the data generated might actually be used to make a difference; and, give scant consideration to the needs of those who might use the data, particularly citizens and citizens’ groups.

In my view, a failure to address these issues risks inadvertently weakening the case for paying attention to, and measuring, aspects of governance. As the Overseas Development Institute’s excellent report on “Governance targets and indicators for post-2015” put it, in diplomatic language: “including something as a target or indicator does not automatically lead to its improvement and the prize is not just to find governance targets and indicators that can be ‘measured’. Rather, it  may be important to reflect on the pathways through which set targets and indicators are thought to lead to better outcomes and on the incentives that might be generated by different measurement approaches.” (See my working document on “Fiscal Governance and Post-2015” for additional thoughts on the inclusion of governance in the post-2015 framework, including notes toward a theory of change).

More broadly, beyond the confines of the post-2015 debate, the risk – and arguably, in many cases, the reality – is that by paying insufficient attention to some key issues, we end up with a lot of data on various aspects of “governance”, but that that data doesn’t get used as much as it might, isn’t very useful for informing context-specific efforts to improve governance, and has limited impact.

To remedy this situation, I’d suggest that any effort to measure aspects of “governance” or to improve the availability, quality, use and impact of governance data (as the Governance Data Alliance is doing – with a Working Group on Problem Statements and Theories of Change) should answer up-front a series of simple questions:

• Outcomes: What outcome(s) are you interested in? Are you interested in improving governance for its own sake, because you regard a particular type of governance as intrinsically valuable, and/or because you think, for instance, that improving governance will help to improve service delivery and accelerate progress against poverty? (See Nathaniel Heller’s post on “outputs versus outcomes in open government”)

• Theory: If your interest is not solely based on the intrinsic value you attach to “governance”, which aspects of “governance” do you think matter in terms of the outcomes – e.g. service delivery and/or reduced poverty – that you’re interested in? What’s the theory of change that links governance to development outcomes? Without such a theory, it’s difficult to decide what to measure!

• Data: In what ways do you think that data about the aspects of governance that you think are important – for intrinsic or extrinsic reasons – will be used to help to drive progress towards the type of governance that you value? To what use might the data be put, by whom, to do what? Or, from the perspective of data-users, what information do they need to take action to improve governance?

Organizations that are involved in generating governance data no doubt spend time considering these questions. But nonetheless, I think there would be value in making that thinking – and information about whether and how the data gets used, and with what effect – explicit.

Addressing these questions explicitly should – by informing decisions about what to measure, how to measure, and how to make the data available – help to ensure that the data produced about various aspects of governance is useful, gets used and has an impact. That’s the point of measuring governance.

Thanks to Matt Andrews, Olimar Maisonet-Guzman, Diarmid O’Sullivan, Andrew Palmer and Leni Wild for continued conversations that have informed this post and to colleagues at Global Integrity for comments.

Photo Credit: antony_mayfield (Flickr: Creative Commons Attribution 2.0 Generic)