Getting beyond the “every country is unique” mantra


This page in:

Moving away from ‘best practice’ thinking has profound implications for development policy work. The craft of policymaking is not simply about delineating the desirable: it is about finding entry points that are both feasible and value-adding. How is this to be done? After all, practitioners desire more guidance than the simple dictum that the answer is ‘country-specific’.

This blog post introduces a dynamic typology (a trajectories framework) to facilitate thinking about the interactions between governance and development policymaking. [A more in-depth introduction is available here; for a more comprehensive discussion, see the paper I co-authored with Francis Fukuyama.] The typology comprises five distinct categories – with each characterized by a distinctive set of governance related opportunities, constraints, and risks.

Category #1: The “founding” political settlement.  ‘Political settlements’ – bargains among elites (and sometimes their non-elite allies as well) that end violent conflict – are key to enabling conflict-affected countries to begin a new chapter in their development histories.

The analysis of “political settlements” is a vast and difficult subject in its own right. Mushtaq Khan’s ongoing work is pathbreaking ; the subject will also be addressed comprehensively  in the World Bank’s forthcoming 2011 World Development Report.  The key point here is simply that the character of this “founding” settlement sets a country along one of two broad trajectories: The resulting political order could be organized around a dominant political party or leader; alternatively, it could be more competitive. Further, within each trajectory the challenges are very different for countries in an early-stage than in the later stages – making for four distinct categories. 

Category #2: Early-stage state dominance incorporates countries where dominant political leadership (perhaps military, perhaps civilian; perhaps organized around a political party, perhaps a charismatic individual) has successfully consolidated its grip on power.

Korea in the 1960s and early 1970s, and contemporary Ethiopia and Rwanda, illustrate this category. In the first flush of authority, there can be a general sense of a country on the move – especially if the leadership has embraced systematically a coherent, and seemingly technically well grounded, strategy for development. However, there are significant risks of reversal – leadership could begin to use its authority for narrowly self-seeking ends, and/or could lose legitimacy (which could range from enthusiastic support to resigned acquiescence) of the broader population.

Category #3: Later-stage state dominance.  Countries that successfully traverse the earlier-stages of the ‘dominant state’ trajectory will over time increasingly confront a new generation of governance-related challenges.

Success in accelerating economic growth generally will have resulted in a more sophisticated private sector, a growing middle class, and an emerging network of civil society organizations – all of which will be seeking institutions capable of both the impartial resolution of disputes, and the provision of a platform for ‘level-playing-field’ competition. This enhanced social and economic complexity is likely to come into increasing conflict with control-oriented political and state institutions – with the outcome uncertain.  Tunisia’s current uncertainties illustrate the challenge; Korea’s successful transition to democracy since 1987 points towards the opportunity.

Category #4: Early-stage competitive clientelism is likely to be the prevalent pattern in those settings where the initial political settlement centered around a ‘founding’ election.

Commonly, successful founding elections translate into a relatively open political culture. The dilemma, though is that, after the flush of electoral enthusiasm has worn off, institutions capable of supporting continuing competition are unlikely to be in place. The result, as in Bangladesh and Kenya, can be a polity in which the rules of the game are clientelistic – and which teeters, seemingly indefinitely, on the edge of chaos. Interestingly, it turns out that this ‘edge of chaos’ can be compatible with significant economic dynamism. But whether a country can sustain its high-wire act for a sufficiently long time for these gains to be consolidated – or whether the ‘edge of chaos’ is a prelude to  backsliding is uncertain.

Category #5: Later-stage competitive polities can emerge via two distinct routes.  They might emerge out of early stage ‘competitive clientelism (although it is troublingly difficult to find examples in history that have managed a sustained evolution – open throughout – from an early to a  later-stage competitive polity). Alternatively, formerly dominant state polities might become competitive polities via a trajectory shift – a ‘zig-zag’.

The ‘zig-zag’ could proceed relatively smoothly – as in Korea, where ‘later stage state dominance’ evolved fairly seamlessly into a competitive polity. But it could also unfold more discontinuously – as in Indonesia and Thailand, where electoral competition served as a way out of a ‘stuck’ state dominance path. Insofar as the political opening of hitherto dominant state polities unleashes previously suppressed conflicts, the governance challenges are likely to be profound. The result could be an ‘edge of chaos’ set of challenges uncomfortably similar – notwithstanding substantially higher levels of per-capita income to countries in the ‘earlier stage competitive’ category.

Keep in mind that a typology is a conceptual construct. I have no intention to suggest that, by grouping countries into categories, one can summarize who whole of any country’s development evolution. The point is not to replace “one size fits all”, faith-based development prescription, with a simplistic “four sizes fits all” pluralism. As my contribution to the October 2010 Journal of Democracy symposium on democratization and development – “The Case for Principled Agnosticism” – makes clear, we need to keep our minds open. But I am hopeful that the distinctions among the categories are sufficiently vivid that they can help us move beyond the analytical defeatism implied by “every country is unique”. What do you think?



Brian Levy

Professor at SAIS and University of Cape Town

Join the Conversation

Alan Hudson
January 27, 2011

Categorising countries/situations certainly seems to provide a way forward.

But the question is whether the typology then provides any additional guidance as to what external actors (including donors) can most usefully to do improve governance in those sorts of countries/situations.

I have some doubts and your blog entry doesn't really allay those. However, a quick glance at the papers suggests that they do go further along the providing guidance path. I shall dig deeper into the longer papers that you mention!

Brian Levy
January 28, 2011

Alan Hudson's useful comment underscores that typologies are only useful insofar as they help differentiate among actions -- how do specific priorities differ across different settings?

Indeed, the more detailed papers Alan refers to offer some guidance. To illustrate, here is one specific example:

How might the priority and potential of public sector management interventions differ across the different categories? The typology helps extend the insights in Phil Keefer's earlier post on "political parties and public sector reform":….
- public management (as distinct from accountability) reforms are likely to hold promise, as both feasible and as high-return priorities, in "early-stage state dominance" settings, and in "later stage competitive polities". In the former case, they are well-aligned with the putative priorites of government; in the latter case, political discourse is (hopefully) beginning to move towards programmatic concerns of performance which are dear to an emergent middle class;
- in "later-stage state dominance" settings, the priority will/should have shifted away from a narrow focus on public sector capacity to a much broader focus on "non-executive institutions of accountability". [The longer these efforts have lagged, the greater is likely to be the urgency of making this shift........]
- in "early stage competitive clientelist settings", it is likely to be unusually difficult to find "handholds" (actors/incentives) which can provide the traction needed for efforts to strengthen public management to move forward credibly. A focus on "islands of effectiveness" is likely to yield greater dividends....

All of the above is, of course, subject to more detailed empirical work. But this is the strength of a typology: it offers a framework within which disaggregated, contestable propositions as how to prioritize can be explored.

Sue Unsworth
February 01, 2011

I agree that it's important to get beyond the "every country unique" mantra and to look for commonalities across countries. I also agree that some basic typology is a useful starting point for understanding what kind of state/political system one is dealing with and the implications of that for priorities, entry points and so forth. So it is essential to differentiate between "collapsed" states and states where there is some established political authority; and between "early-stage" and "later stage" countries where different levels of institutionalisation (within the state, the private sector and civil society) and different income levels will fundamentally affect incentives and capacity for development. These categories would roughly align with North's "fragile", "basic" and "mature" limited access orders, I think.

I also agree that it’s useful to consider the impact of different levels of political competition and how that shapes the incentives of political elites and their scope for action. So the broad categories that Brian suggests are a good place to start, especially for agencies such as the World Bank and the European Commission that work in both low income and middle income countries. (I do have one quibble with the differentiation between categories 2 and 4 which is that rules of the game are clientelistic in all "early-stage" low income countries, and also that it is the nature and extent of political competition as shaped by the political settlement, not just the formal electoral system, that is important).

But I wonder how far these categories provide a useful guide for action. For one thing, as Brian recognises in his May 2010 article, most governance arrangements are hybrids and countries may not stay on the same trajectory. Moreover, most low income countries which pose the greatest challenge for donors fall into category 4 (early-stage competitive clientelism). So the pressing need is for more guidance on understanding how and why different clientelist systems work differently in different country contexts within category 4. The nature of the political settlement is clearly an important variable in helping to explain differences in more/less effective management of rents between countries and over time within one country (for example in Kenya between Kenyatta and Moi; or in Uganda at different times under Museveni). But other factors are also important including social and institutional legacies, sources of government revenue (including aid), the external environment and so on (and the last two may be actionable by external players).

Within category 4 countries it seems particularly important to try to understand local political dynamics – including at a subnational level – partly because there is considerable scope for external actors to do harm in such contexts, and partly because progress seems to depend on looking for ways to support incremental change – perhaps by building on existing local structures, or nurturing whatever scope there may be for productive bargaining between elites (for example between politicians and investors with shared interests in economic growth; or between revenue authorities and taxpayers). Thinking about how the global or regional environment creates positive or negative incentives for local elites (the "global drivers" agenda) also seems particularly important in category 4 countries.

The ODI's Africa Power and Politics Programme is aiming both to increase our understanding of how political systems work including at a very local level, and to look for commonalities across countries. Some of the emerging findings are quite sobering for donors including the damage done by constantly changing donor fashions and the potential for programmes designed to “strengthen” civil society instead to weaken incentives for collective action and to increase inequality and corruption. We need more of this kind of research.

Brian Levy
February 08, 2011

Sue Unsworth, and others, raise an interesting challenge as to the value added of a typology -- namely the extent to which it can offer a "guide to action". As my blogpost suggests, we somehow need to navigate the space between 'every country is unique' and 'four sizes fit all'. Two thoughts on this:

First, as the global economy's near-death experience with the recent financial crisis has taught us, "black swan" events are out there. Even as social science modellers (and dynamic typology builders.....) work to predict 'normal' patterns as to how events unfold, the unexpected also can happen. Put differently, it is helpful to distinguish between, on the one hand, the normal pattern of evolution along each of the trajectories and, on the other, discontinuous zig-zag leaps across trajectories. Ordinarily, action is likely to be oriented towards moving forward within a trajectory -- although sometimes, where things seem seriously stuck, there might be a case for seeking to foster a zig-zag.

Second, in part (but not only) because of the above, my sense is that the implications of locating a country within 'typological space' are best thought of in probabilistic terms. Without closing any doors, a typology offers a guidepost as to which potential entry points for action are worth exploring further, and which (without rejecting entirely the possibility of success) are better viewed sceptically. A few examples:

- differentiating (as per the exchange with Andrew Hudson above) the kinds of settings where public sector management reforms are likely to get traction, and where they are not;

- shifting focus away from national-level and towards local-level entry points -- a shift which, precisely along the lines suggested by Sue, seems most relevant for category 4 countries;

- assessing the urgency to be given to efforts to strengthen non-executive institutions of accountability -- a process which need not be on the front-burner for category 2 (early stage state dominance) countries, but which they delay too long at their peril (as Tunisia and Egypt are in the process of demonstrating vividly).

To be sure, there can be no certainties -- at least not until we have research which shows that "interventions of type X never work in category Y settings" ......and/or sometimes/always work in category Z settings". I sincerely hope that research will help narrow the range of uncertainty.

But I also expect -- and hope -- that uncertainty will persist. For it is uncertainty itself that engenders two assets that, more than any amount of knowledge, are crucial to effective engagement with the challenge of development -- a learning orientation, coupled with humility as to the limits of how much we can know.

Alan Hudson
February 08, 2011


I'm glad that you've taken the opportunity to spell out in more detail the ways in which you think that your categorisation/typology can be useful. I think that your initial blog-entry sold your typology a bit short.

As you say, the challenge is to navigate the terrain between advocating blueprints and repeating mantras of context specificity. Your four-sizes fits-all recognises that challenge and in attempting to address it is a real step forward. I've not come across other stuff that grasps this nettle as effectively as you have.

However, your latest comment raises in my mind some more fundamental questions about the role that external actors such as the World Bank can effectively [I will leave legitimacy aside for the moment, but others may not!] play in enhancing governance in developing countries. It would be interesting to get a sense of what discussions there have been at the Bank on this - let's hope that the new-found enthusiasm for transparency extends this far ;-)

Put differently, in your first para in your comment directly above, who is the "guide to action" for, who acts and what role should external actors play?

I haven't given up on the idea that external actors such as the Bank and DFID can play a role of sorts in enhancing governance, but I do sometimes think that it needs to be much more realistic about its limits. Perhaps nurturing an environment of transparency and accountability out of which local solutions can emerge might be the best - most effective and most legitimate - place to draw the line?

Alan - working for DFID, but writing in a personal capacity