What does good governance mean?

Gisselquist / Jan 2012

Almost all major development institutions today say that promoting good governance is an important part of their agendas. The outcome document of the recent 2011 Busan High-Level Forum on Aid Effectiveness further reflects these commitments. In a well-cited quote, former UN Secretary-General Kofi Annan noted that ‘good governance is perhaps the single most important factor in eradicating poverty and promoting development’.

Despite this consensus, ‘good governance’ is an extremely elusive objective. It means different things to different organizations, not to mention to different actors within these organizations (to make matters even more confusing, governance experts also routinely focus on other types of governance—global governance, corporate governance, IT governance, participatory governance, and so on—which may be related only peripherally to the good governance agenda vis-à-vis domestic politics and administration which is our focus here.)

Good governance as a concept

In general, work by the World Bank and other multilateral development banks on good governance addresses economic institutions and public sector management, including transparency and accountability, regulatory reform, and public sector skills and leadership. Other organizations, like the United Nations, European Commission, and OECD, are more likely to highlight democratic governance and human rights, aspects of political governance avoided by the Bank. Some of the many issues that are treated under the governance programmes of various donors include election monitoring, political party support, combating corruption, building independent judiciaries, security sector reform, improved service delivery, transparency of government accounts, decentralization, civil and political rights, government responsiveness and ‘forward vision’, and the stability of the regulatory environment for private sector activities (including price systems, exchange regimes, and banking systems).

In short, working uses of the term ‘good governance’ include a variety of generally ‘good’ things. But these ‘good’ things do not necessarily fit together in any meaningful way. Indeed, good governance would be a great example of a poorly specified concept for an introductory course in social science methodology. What makes a concept good? In a 1999 article, political scientist John Gerring spelled out eight ‘criteria of conceptual goodness’ that provide a useful framework. Four of these criteria are especially relevant here:

  • First, ‘good governance’ lacks parsimony. Unlike good concepts, good governance has endless definitions, and we always need the details of each to understand if we are talking about the same thing.
  • Second, ‘good governance’ lacks differentiation. Well-governed countries often sound a lot like functioning liberal democracies, for instance, and it is not clear how they differ.
  • Third, ‘good governance’ lacks coherence. Its many possible characteristics—from respect for human rights to efficient banking regulations—do not clearly belong together.
  • Fourth, and most important, ‘good governance’ lacks theoretical utility. It confuses, rather than aids, in the formulation of theory and the related project of hypothesis testing, not least because the concept is so fluid that analysts can easily define it in the way that best fits their data.

Methodological discussions are often esoteric and best kept within scholarly circles, but this one has real world relevance to development policy. Donor agencies regularly measure and assess the quality of governance, and may condition assistance on these measurements. The Millennium Challenge Corporation of course is one of the most explicit in doing so, but it is in good company. Donors also purport to design and implement evidence-based policies on governance reform. They further justify this focus on good governance partly on the basis of evidence that better governance promotes economic development.


The weakness of the good governance concept, however, calls into question each of these projects. Without stronger concepts, donor agencies have no clear basis upon which to argue the merits of one measurement versus another or to evaluate the relative importance of various components of governance in any classification. Without better measures, donor agencies cannot, in a rigorous manner, empirically test hypotheses about how political and economic institutions change, much less develop evidence-based strategies about how to positively influence this change. Nor can they be very convincing about the rigor of quantitative findings suggesting a causal relationship between (weakly-conceptualized) measures of governance and development outcomes.

Rwanda provides just one illustration of some of these issues. As many observers note, Rwanda has made clear progress in terms of economic growth, public sector management, and regulatory reform since the genocide in 1994. As many other observers note, its record with respect to democracy and respect for civil and political rights has been extremely problematic. Should Rwanda be considered well governed because of its economic progress, or poorly governed because of its democratic deficits? The UK’s Department for International Development (DFID), for instance, has been the largest bilateral donor in Rwanda, effectively arguing the former. Human Rights Watch, among others, sharply criticized DFID policy last year, effectively arguing the latter.

The question of ‘how to improve governance?’ is of course the most pressing from a policy perspective. However, this question cannot be rigorously answered without better addressing the concept of good governance: ‘how to improve what exactly?’. These points are discussed in greater depth in my forthcoming working paper, ‘Good Governance as a Concept, and Why This Matters for Development Policy’. The paper provides a review of donor approaches to governance, discusses conceptual issues in greater depth, and argues that one promising way forward is to disaggregate the concept of ‘good governance’ and to refocus our attention and analysis on its various disaggregated components (e.g., democracy, civil and political rights, public sector management). In short, the term ‘good governance’ has become a catchy shorthand way to describe a variety of political and economic institutions and outcomes and is thus likely to remain in common public usage, but, as is, it is not a useful concept for development analysts and policy makers.

Rachel M. Gisselquist is a Research Fellow with UNU-WIDER.

This article is based on UNU-WIDER working paper no. 2012/30, 'Good Governance as a Concept, and Why This Matters for Development Policy'. It was originally appeared in the January 2012 issue of the WIDERAngle newsletter.