Monday, November 30, 2015

Modeling Fragility: A Societal and Institutional Approach, by Seth KAPLAN

The OCED is trying to build a model to better analyze fragility. This is admirable, but also risky. Past attempts to model fragility—such as the Fragile States Index (FSI)—have done more harm than good and I wonder whether no model would be better than another model based on misconceptions.The danger of focusing on what’s measurable

In particular, I worry that the need to produce something quantifiable and useful for donors will lead to an overemphasis on the measurable issues that matter most to those in the North, such as violence and security. Other issues – say, societal cleavages or inequitable institutions – that cannot be easily measured or that require deep engagement with local societies will be left out or undervalued.

The models we’ve used so far focus on issues that have no causal relationship with fragility (such as population growth and income levels); are products, rather than causes, of fragility (such as violence and corruption levels); or are based on Western political norms (such as regime type).

If it was up to me, I would develop a model that focused on the fundamental drivers of fragility.

Zooming in on societal and institutional dynamics

Societal and institutional dynamics frame how more formal institutions and processes work. They determine the quality of government, the inclusiveness of economic and political systems, and strength of the centripetal or centrifugal forces acting on society. They influence or even determine the factors – violence, resilience, institutions, justice, economy – that appeared in last year’s OECD report.

A model built on the dynamics of institutions and society would require qualitative assessment. While this would not readily provide the lists and measurements favored by international actors, it would zero in on the issues that most affect SDG 16 and the PSGs.

The importance of social cohesion

Social cohesion is especially important in less developed countries because formal institutions are weak and often susceptible to manipulation, corruption, and bias. Unlike their more institutionalized brethren in the developed world, these states feature formal institutions incapable of neutral mediation and enforcement of rules – unable to deliver truly public goods. As a result, elites and officials have too much discretion to bend the rules and appropriate the resources of the state.

When formal institutions are weak, social cohesion can, to a certain extent, encourage leaders to resolve problems amicably and with a public spirit, as has happened at crucial points in Somaliland, Chile, and Tunisia, for instance. Without social cohesion, it is very hard to improve formal institutions—the approach typically advocated by donors—because elites and officials accustomed to putting their own interests before the public good have strong incentives to undermine such reform.

On the other hand, if a state is strongly institutionalized, these social fractures matter much less. Governments underpinned by strong autonomous institutions are more likely to observe the principle of neutrality and be much fairer managers of conflict and distributors of resources.

Seen this way, fragility can be understood as existing along two dimensions. Low institutionalization and low social cohesion are at one corner (occupied by countries such as Somalia, the Central African Republic, and South Sudan), and cohesive, highly institutionalized nation states are diagonally opposite.

Four Types of Political Orders (with examples)[i]

Low Political-Identity Fragmentation
High Political-Identity Fragmentation

High Institutionalization (or at least high coercive capacity)
I: Dynamic


III: Fragile but Controlled

Syria (before 2011)
Soviet Union
Iraq (before 2003)
Saudi Arabia

Low Institutionalization
II: Stable but Sluggish

IV: Fragile and Unstable

Libya (after 2011)
Syria (after 2011)

States towards the fragile corner are trapped in a vicious cycle of societal fragmentation and weak institutions that feed on one another and make escape difficult. The combination of rigid social divisions and weak state institutions in Lebanon, Libya, and Yemen, for instance, mean that institutions become arenas for sometimes violent competition over power. In African countries such as Nigeria and Kenya, the state lacks the autonomy and capacity to manage conflict and drive development forward.

To identify and assess fragile states, it is essential to break down as many of the societal and institutional sources of fragility—the places where societal cleavages are created or exacerbated—as possible. This approach would be far more comprehensive and more keenly focused on key dynamics. It does not conflate resilience (or luck) with true robustness; states would not be characterized as “non-fragile” when, in fact, they sit atop combustible societies (Syria before 2011). Existing indices and tools have repeatedly proved poor predictors of conflict or state failure; many of the Arab countries now in turmoil (for instance, Libya and Bahrain) did not make fragile states lists before 2011.

It would clearly be better to flag potentially fragile states than identify them after the event. These potential fragile states then would need to be comprehensively assessed to determine their true nature and possible counter measures. There are three relatively simple clusters of indicators that could approximate some of the fundamental societal and institutional dynamics and each could be used in conjunction with other indicators of fragility.

·         Ethnolinguistic diversity: research shows that greater ethnolinguistic diversity produces greater conflict, lower growth, and inferior development outcomes because it sets up the conditions for greater animosity, lower trust, higher transaction costs, and reduced demand for public goods and redistribution. This could be paired with an indicator such as institutional robustness.

·         Horizontal inequalities (HIs): this indicator would build on the work of Frances Stewart and others that documents the role of HIs in conflict. It can be based on a cluster of Gini coefficients (computed using disaggregated data for groups, regions, or income levels) for a wide range of public services.

·         Perceptions: this indicator would partly build on the work done by the International Dialogue for Peacebuilding and Statebuilding (IDPS), which has recommended using perception surveys in fragility assessments. Surveys could examine differences across disaggregated data for groups, regions, or income levels for a cluster of issues including representativeness and effectiveness of public services; confidence in the performance of informal/formal judicial institutions; intergroup relations; cross-group cooperation; economic conditions; safety and security; performance of security institutions; confidence in the future; and whether the benefits of natural resources are shared equally.

Of course, it is important to remember that change ultimately depends on how indigenous actors conduct themselves and treat each other. Government is the best mechanism for bringing actors together and building national consensus, which is why the New Deal puts government at the center of its process, but social institutions can also play a bridging role.

International actors should concentrate on catalyzing domestic processes. They may be better placed to act in the short term, but long-term impact can only come from investment that addresses problems such as horizontal inequalities and economic imbalances. Any new fragile states model must be able to show us clearly where our efforts are best directed.

The views expressed in our blogs remain those of the authors and do not necessarily represent the views or policies of the OECD or its members

Seth D. Kaplan, a Professorial Lecturer in the Paul H. Nitze School of Advanced International Studies (SAIS) at Johns Hopkins University, has written two books on fragile states, Fixing Fragile States: A New Paradigm for Development and Betrayed: Promoting Inclusive Development in Fragile States, and runs the website

[i] A close examination of both quantitative data and qualitative case studies comparing comparable countries within regions produced this result. See my book, Fixing Fragile States, to learn more.