Document Library

Key Text Unbundling Institutions at the Local Level: Conflict, Institutions and Income in Burundi

Author: Maarten J. Voors, Erwin H. Bulte
Date: 2008
Size: 44 pages (152 KB)

Access full text: available online

Summary

What kinds of institutions most effectively boost personal income and reduce inequality? This paper, published by the Institute of Development Studies, identifies institutions that guarantee property rights security as the most vital drivers of long-term prosperity. Insights from a new data set from Burundi reflect the comparatively less important roles of local political leadership and social capital. Conflict plays a positive role in income growth, largely because conflicts may contribute to institutional maturation.

A number of explanations have been proposed to explain income differentials, including differences in capital accumulation, technical expertise and suitable market conditions. According to an emerging consensus, institutional development is a major determinant of income growth across all societies. This concern for institutions usually includes both formal and informal rules of structured social interaction, referring to cultural practices as much as to governing bodies.

This study of conflict, institutions and income in Burundi, carefully measured and quantified, yielded the following findings:

In confirmation of results from cross-country studies, there was an unmistakable causal link from secure property rights to income.
Once the variable of property rights security was controlled for, political institutions were not directly relevant to raising income levels.
Social capital also appeared not to play a significant role in promoting growth in Burundi.
Political leadership and social capital may make a positive contribution to development, indirectly. Examples would be local political authorities impacting the distribution of income, or social capital networks playing a role in insurance.
In a reduced-form regression looking at income and armed attacks, either by rebels or the army, there is actually a positive correlation between conflict and income. Conflict intensity may strengthen local institutions, thus promoting economic growth.

This analysis, both in its method and in its results, suggests a number of implications for scholars working in this field and for policy-makers seeking to facilitate economic growth:

A cross-section analysis undertaken at the micro-level, with community and household data, enabled a close look at within-country inequality. For a granular picture of local institutions, other studies should also scrutinise income, demographic and tenure data from many households in a random sample.
A three-fold distinction made between types of institutions - as opposed to a narrow look only at those which protect individual property rights - was vital for capturing the larger institutional framework. The model applied here added political institutions and social capital mechanisms to property rights. Such distinctions, new to this study, can aid policy-makers more than general recommendations from cross-country reports.
The novel instrument of a series of 2SLS regressions may be useful for future micro-level research in conflict situations. These analyses demonstrated that much of the violence that occurred in Burundi was indiscriminate, and measuring levels of conflict intensity cast further light on variable institutional performance.
The results suggest that both formal and informal tenure security may promote income growth. Incentivising citizens to invest in land productivity and erosion control was decisive.
The promotion of land titling, already underway in Burundi, Rwanda and Ethiopia, appears to be another pragmatic and workable means of enhancing rural incomes.

Access full text: available online

Source: Voors, M.J. and Bulte, E.H., 2008, 'Unbundling Institutions at the Local Level: Conflict, Institutions and Income in Burundi', Households in Conflict Network, Institute of Development Studies, Brighton
Author: Maarten Voors , maarten.voors@wur.nl
Organisation: Households in Conflict Network (HICN), http://www.hicn.org/index.html