Survey of traditional and political village chiefs in Uganda: Institutions, land tenure, and water provision.

The dataset is a survey of 300 village chiefs in Uganda, equally split between traditional (chiefs, clan leaders and elders) and political authorities. The respondents were randomly selected based on a stratification that is representative of the largest 12 ethnic groups, i.e. about 85% of the Ugandan population. The data is GPS coded and can be linked to larger socio-economic datasets such as the LSMS-ISA. It contains information on the relative power of formal political versus informal traditional authority regarding the resolution of social and economic conflicts. It also contains information on practices and attitudes regarding land tenure (property rights, trans- actions, and prices) as well as water provision (initiatives for irrigation and drinking water). The data is useful for studying the impact of institutional arrangements on social and economic outcomes.

Documentation on dataset: [chief_dataset_description]

Questionnaire of the survey: [UGA_leader_survey_questionnaire]

To access the dataset, please write to or


Focus Paper: Communal Land and Agricultural Productivity

In most countries, labor productivity is lower in agriculture than in the non-agricultural sector. This agricultural productivity gap (APG), moreover, is substantially larger in developing countries, as can be observed from Figure 1. For example, the average APG in the poorest decile of countries is 7.8 higher than in the richest decile of countries. This enormous difference is a key piece in the development puzzle given that farming accounts for 73% of employment in the poorest economies. Is their low agricultural productivity the result of any particular policies? And, if yes, do these policies account for a major loss in aggregate output?

Focus paper can be downloaded here [pdf].

Communal Land and Agricultural Productivty

Our first paper on land tenure systems in Africa, in particular Ethiopia, and their effects on agricultural productivity came out as a working paper at the Centre for Macroeconomics.

For policy makers and journalists – [link to executive summary]

For academics – [link to academic paper]

This paper quantifies the aggregate impact of communal land tenure arrangements that prevail in Sub-Saharan Africa. Such tenure regimes limit land transferability by prohibiting sales, subjecting rented-out land to the risk of expropriation, and redistributing it to existing farmers in a progressive fashion. We use a general equilibrium two-sector selection model featuring agents heterogeneous in skills to compute the resulting occupational and operational choices as well as land allocations. The quantification of the model is based on policies deduced from Ethiopia. In the Sub-Saharan African context we find that such policies substantially dampen nominal agricultural relative to non-agricultural productivity, by 25%. Real relative agricultural productivity, however, only falls by 4% since cross-sectoral terms of trade adjust strongly, with excess agricultural employment only amounting to some 1.5 percentage points. The loss in GDP is small, about 2%. That serves as a reminder that ostensibly highly distortionary policies need not have substantial bite when individuals strategically adjust to them and equilibrium prices adapt. For example, the model predicts that at given prices 62% of farmers in an economy such as Ethiopia would leave farming if tenure were secured, casting land insecurity as a major obstacle. Yet only 9% would actually switch sectors after price adjustments are factored in.