Landslide Risk and Risk Reduction: A Quantitative Outlook on a Socio-Environmental Issue in Uganda

I adopt a quantitative approach to measure the causes and consequences of landslide risk in the Rwenzori mountains in Uganda. I thereby aim at contributing to a better understanding of the causal pathways that lead to the reproduction of risks in the region. The first chapter sets the scene. It introduces the research concepts and highlights relevant problems in micro-economic literature. I start by presenting the view that disaster risk is the result of an interaction between (natural) hazards and (societal) processes that lead to exposure and vulnerability. I then highlight how micro-economic approaches can contribute to measuring landslide impact and to identifying pathways that lead to risk exposure and vulnerability. Relevant problems with the core assumptions of micro-economics and empirical caveats in existing literature are thereby discussed. From these caveats I subsequently develop the four main research questions of this manuscript. The first chapter ends with an introduction of the study area, the various types of data and my methodology. I thereby shortly discuss the advantages and disadvantages of using a quantitative approach. In chapter two, I apply an econometric analysis on data from an original cross-sectional household survey in order to measure the direct impact of landslides on households' income and livelihoods. This first, rather basic, accounting of landslide impact is necessary to confirm the relevance and importance of this manuscript. This is the first quantitative study of the socio-economic impact of landslides in a remote area of the Global South. The findings in chapter 2 confirm that it is socially relevant to study landslide risk in Uganda. In the following chapters I turn towards the underlying causes of landslide risk in the Rwenzori region. In each chapter I thereby put into question a different assumption of micro-economic theory. Chapter three investigates why some farmers who are highly exposed to landslide risk do not have the intention to plant trees against landslides, even if they are well aware of the landslide risk. I do thereby not assume that farmers behave in an economically rational way. I argue that low levels of self-efficacy form an internal constraint that is preventing some farmers from taking measures against landslides. Alternative explanations and limitations of my analysis are discussed. In an attempt to understand how exposure to landslide susceptibility is determined, chapter four introduces land markets and land transactions into the analysis. To do this, a cross-sectional dataset on households and plots is turned into a panel dataset on plot ownership. The purpose is to measure the evolution in land ownership over time for farmers who are exposed to landslides and for farmers who are not. I thereby demonstrate that exposure to landslide susceptibility is not exogenous to farmer characteristics, since farmers consciously decide where to buy land, depending on their initial land holdings. While I assume a rational behaviour here, my conceptual framework is based on a theory on risk preferences that is not following the rules of standard utility theory. I investigate both prospect theory risk preferences and disaster avoidance behaviour and find that the latter is the most suitable framework for the interpretation of my results. In chapter five, I expand my set of theories and tools to understand land transactions in the presence of landslide risk. Social norms, rather than individual risk preferences, are put forward to explain transaction patterns on the land market. I make use of a discrete choice experiment to test for the presence of such norms and the importance of these norms in guiding behaviour on the land market. As such, in a final attempt to question standard micro-economic theory, I show that my decision to treat markets as being embedded in a wider social context of familial ties and social norms increases the number of possible explanations for transaction patterns on the land market. In the discussion of this dissertation I take a step back and look at the strengths and weaknesses of my different chapters. The advantages and disadvantages of using economic methods to study disasters are discussed. I thereby critically assess my research, adopting two different perspectives. First, I evaluate internal and external validity of my research from an economic perspective. Secondly, I take a step outside the traditional economic framework and look at the consequences of my research for disaster discourse. These two perspectives allow me to identify strengths and weaknesses I would not have seen if I had adopted one perspective alone. General recommendations for policy and research are derived in the final section of my discussion.

Publication year:2018

Accessibility:Open