Managing Humanitarianism

The mismanagement of humanitarian emergencies can have profoundly negative consequences for regional and international security. My dissertation project examines the conditions under which governments strategically facilitate or constrain the delivery of international assistance in response to humanitarian emergencies, including natural disasters and conflicts. I develop a theoretical framework to explain how governments regulate the presence of humanitarians within their borders in order to minimize reputational threats among international donors and investors. The entry of humanitarians into emergency-affected areas often provides them with negative information about government performance that can — if publicly disseminated — affect future government revenues from international sources. To avoid potential losses, governments can deny the existence of emergencies, under-report the needs and number of people affected, delay declaring an emergency, and limit humanitarian organizations’ access to affected areas. Consequently, while governments retain access to foreign revenue flows, they fail to address the needs of vulnerable populations directly affected by emergencies that can spiral into broader security threats both within and across borders.


To assess this theoretical framework, I employ a tripartite research design. I first document global trends in government restrictions on humanitarian actors to show how governments that are particularly dependent on foreign resources, whether measured by aid or trade, are the most likely to impose restrictions, regardless of regime type, emergency type, or history of conflict. Second, I have conducted a global survey of over 500 humanitarian professionals, coupled with in-depth interviews of donor country officials, to corroborate the conditions under which they are most likely to face government interference during emergency operations. Third, after empirically establishing a surprising amount of variation among countries with similar characteristics, I trace government-humanitarian dynamics through a case study of Niger, a least-likely case for the phenomenon of interest. Through interviews with over 100 humanitarian professionals and government officials working in the country and a survey of 400 national and local government officials in Niger, I demonstrate how the government strategically imposes constraints on humanitarian actors when their activities are perceived to threaten the government’s reputation.


Ethnic Marginalization and (Non)Compliance in Public Health Emergencies

with Leonardo R. Arriola

Forthcoming in The Journal of Politics

Read the paper here

Health crises can reveal the inability of governments to induce compliance with policy interventions. While lack of compliance is conventionally attributed to individual mistrust in government, resistance to such interventions is often found clustered among entire communities, particularly in ethnically divided societies. We account for such patterns by explaining how citizens adjust their responses to state authority according to their shared ethnicity with those in power. We assess the effect of ethnicity on citizen compliance with a public health advisory on HIV/AIDs issued by different authority types through a survey-based field experiment in the West African country of Guinea. Members of a politically marginalized ethnic group, the Peul, are significantly less likely to comply with a public health advisory from a national government representative, the president, when compared to local or religious leaders. We show that perceived ethnic discrimination conditions both trust in and compliance with different authority types.


Funded by Innovation for Poverty Action's Peace and Recovery Program.

Local Solutions to Local Problems? Evaluating the Effectiveness of Educational Programs to Reduce Interethnic Tensions and Local Violence

with William Nomikos and Niloufer Siddiqui

We assess the effectiveness of a new EU-funded project to strengthen  social trust and rebuild communal norms of tolerance in  Burkina Faso. Burkina Faso has a strong history of peaceful coexistence among ethnic and religious groups, but in recent years has seen a rise of organized violence by Islamic extremists and a fracturing of old forms of social solidarity. Through a randomized  curricular intervention administered in schools and youth clubs, this study assesses the effectiveness of individual-level (building self-esteem and self-confidence)  and community-level (encouraging social cohesion through emphasizing shared identities) factors on intercommunal trust and the willingness of individuals to rely on non-violent solutions to local disputes. The proposed intervention seeks to educate  adolescents to reinforce identities that are tolerant of others and resilient to radicalization.  In co-ooperation with a local implementing partner, we randomly assign an  education module to teachers and non-school educators in six communes and two cities (Bobo-Dioulasso and Ouagadougou) in Burkina Faso. These educators will deliver the module to 2,400 youth from the ages of 12-25.

Documenting Violence and Promoting Peace in Africa: A Pilot Study for Data Collection in a Conflict-Affected Country

with Leonardo R. Arriola, Brice Bado, Justine Davis and Aila M. Matanock

Funded by the Carnegie Corporation

The lack of accurate locally sourced data is an obstacle to improving knowledge of electoral violence in African countries. Research on this form of violence is often based on secondary sources that limit the level of detail and even the accuracy of associated data. In this context, we propose to implement four data-collection methodologies — civil society monitors, party monitors, crowd-seeding, and random calling — to compare their effectiveness in generating data on various forms of electoral coercion and violence. We will assess the methodologies based on metrics related to event type (e.g., frequency, location, timing) as well as in relation to publicly available violence datasets. For this assessment, the methodologies will be implemented over an eight-month period in the same selected locations in the run-up to the October 2020 presidential elections in Côte d’Ivoire, a country where
elections have been routinely marred by violence since the reestablishment of multiparty politics. Data will be collected for six months before the election and two months afterward. The project’s resulting assessment will identify trade-offs associated with each data collection methodology in terms of validity, accuracy, scalability, and cost.

©2017 by Allison Namias Grossman. Proudly created with