Risks, Insurance, Shocks: Case Study and Experimental Evidence from Colombia
von Stephan Dietrich
Datum der mündl. Prüfung:2013-12-18
Betreuer:PD Dr. Marcela Ibañez Diaz
Gutachter:Prof. Dr. Stephan Klasen
Gutachter:Prof. Dr. Oliver Mußhoff
EnglischIn this dissertation I present novel findings that analyze the effects of insurances on the resilience to shocks and household's risk taking behavior. Moreover, the question is raised as to whether there could actually be something good about shocks, as they could trigger processes of social cohesion. The four chapters of this dissertation focus on two specific forms of insurances including an agricultural insurance and insurance networks exogenously formed in experiments. Risks are a major impediment for households to escaping poverty. Investments that could improve the well-being are declined if failures cause catastrophic consequences. In addition to households own decision to face risks, external shocks such as natural disasters or wars can drag them into chronic poverty. Insurances for poor households have been paid increasing attention in the last decade as a tool to reduce risks. Efficient insurances are expected to encourage the decision maker to take risks and to strengthen their resilience to adverse events. In cases where an insured household suffers a shock, indemnification payments mitigate the financial impacts. As catastrophic outcomes are less likely with an insurance, they encourage households to pursue opportunities promoting investments and enhancing households welfare. The geographical research focus is put on Colombia throughout all chapters. As the most recent publications on this topic focus on Asia (particularly India) and Africa, this work presents findings in a geographical context that has been underrepresented despite it's high need for efficient risk management. Colombia has been exposed to a large range of shocks in the last decades. Firstly, the ongoing civil conflict that has caused displacements of nearly 6 million people and cost the life of more than 177 000 civilians between 1958 and 2012. Furthermore, in 2010 and 2011 the weather phenomenon La Niña resulted in heavy rains and floods causing major damages that affected about 3 million Colombians and resulted in emergency relief programs summing up to about 400 million. US$ in humanitarian aid. The first part of the dissertation focuses on the analysis of the impacts of the agricultural insurance. The analyzed insurance is relatively new and covers weather related losses. The scheme is based on a traditional design in which indemnification payments are triggered upon individual verifications of damages. Essays 1 to 3 are devoted to an evaluation of this insurance program. In the first essay I present the results of the effect of the insurance program on household loans, assets, expenses and income. The special way in which the insurance was offered allows a quasi experimental identification of the program impacts. The second essay investigates the link between ex ante insurance and ex post risk management. More precisely, it analyzes to what extent indemnification payments affected the resilience to shocks. Inconsistencies in the allocation of indemnification payments and a surprisingly low risk reduction associated with the insurance offer the rare opportunity to compare indemnified and non-indemnified households conditional on their losses. The third essay analyzes if purchasing the insurance affected the risk taking behavior of farmers. Particularly, the chapter describes the findings on how the insurance impacts household loans and productive outcomes. The data used in the analysis correspond to a survey that was especially designed to address the research questions. For that purpose data from nearly 500 tobacco producing households were collected in Santander, Colombia. The second part of the dissertation focuses on the effect of shocks on the development of social ties. While adverse shocks cause large human and economic losses, the recent literature emphasizes that something good can result from shocks. Following this intuition, the fourth essay aims at investigating how the exposure to shocks affects social cohesion. In particular, it is considered whether the frequency and covariance of shocks affect the development of social ties in informal insurance networks. To account for the endogenous formation of networks and to get control of the structure of shocks, laboratory experiments were implemented. The experiments were conducted in the computer lab of the Universidad Nacional in Medellin, Colombia, with 540 students.
Keywords: Risk Management; Insurance; Impact Evaluation; Experiments; Solidarity Networks;