This thesis argues that King Leopold II, in his exploitation of the Congo, dealt the Congo a future of political, ethnic, and economic destabilization. At one time consisting of unified and advanced kingdoms, the Congo turned to one completely beleaguered by poverty and political oppression. Leopold acquired the Congo through unethical means and thus took the people's chances away at self-rule. He provided for no education or vocational training, which would stunt future Congolese leaders from making sound economic and political policies. Leopold also exploited the Congo with the help of concession companies, both of which used forced labor to extract valuable resources. Millions of Congolese died and the Congo itself became indebted through Belgian loans that were given with no assurance they could ever truly be paid back due to the crippled economy of the Congo. With the Congo now in crippling debt, the current president, Joseph Kabila, has little incentive to invest in reforms or public infrastructure, which stunts economic growth.1 For over a century the Congo has been ruled by exploitative and authoritarian regimes due to Leopold's initial acquisition. The colonization from Leopold lasted from 1885-1908, and then he sold it to his home country of Belgium who ruled the Congo from 1908 to 1960. Belgium helped prop up a dictator named Joseph Mobutu or Mobutu Sese Seko who ruled from 1965 to 1997. Afterwards he was overthrown by the Kabila family who has continued the exploitative rule and has made no significant efforts at democratization or reforms. Thus the ethnic conflicts, political oppression and economic woes that the Congo is facing today are inevitably linked to its Leopoldian past.

Internationale Geldgeber haben die Finanzierung für den Schutz tropischer Wälder erhöht, um der globalen Herausforderung von Klima, Biodiversität und Nachhaltigkeit zu begegnen. Allerdings fehlen subnationale Informationen darüber, wo und wie die Gelder verteilt werden, welche Faktoren die Finanzierung beeinflussen und wie sie mit der Dynamik der Wälder und geschützten Gebiete korrelieren. Diese Thesis beabsichtigt, diese Fragen zu beantworten, indem sie sich mit drei Jahrzehnten internationaler Naturschutzfinanzierung in den Hauptabholzungsgebieten Südamerikas auseinandersetzt. Mithilfe gemischter Methoden habe ich die Interessen der Geldgeber thematisch und geografisch kartiert, räumliche Determinanten der Mittelvergabe identifiziert und Schwankungen der Finanzierung über Standorte und Zeit hinweg mit der Dynamik der Waldbedeckung und geschützten Gebiete verknüpft. Die Ergebnisse zeigten, dass die internationale Naturschutzfinanzierung eine Mischung aus global relevanten Interessen und bilateralen Interessen darstellt, ermöglicht durch sozioökonomische und biophysikalische Verbindungen zwischen den spendenden und empfangenden Regionen. Trockenwaldökosysteme mit hoher Abholzungsrate waren besonders unterfinanziert und gefährdeten die Ökosysteme, Arten und die lokale Bevölkerung. Die Verknüpfung von Schutzgebieten und Finanzierung mit dem Konzept der Landnutzungsdynamik enthüllte weitere Nuancen und half, kontextspezifische Empfehlungen zu identifizieren. Diese Studie präsentierte die erste subnationale Analyse der internationalen Naturschutzfinanzierung auf kontinentaler Ebene, zeigte Übereinstimmungen und Diskrepanzen zwischen den zugeteilten Ressourcen und den Naturschutzbedürfnissen und beleuchtete die komplexe und dynamische Landschaft der Finanzierungsmöglichkeiten, mit der andere Akteure umgehen müssen. / International donors have increased funding for tropical forest conservation to address the global challenge of climate, biodiversity, and sustainability. However, subnational information on where and how funds are allocated, factors influencing funding, and its correlation with forest dynamics and protected areas is lacking. This thesis aims to answer these questions by delving into three decades of international conservation funding in South America’s major deforestation areas. Using mixed methods, I mapped donor interests thematically and geographically, identify spatial determinants of funding allocation, and link funding variations across locations and time with forest cover and protected areas dynamics. Results found that international conservation funding carried a mix of globally relevant interests and bilateral interests enabled by socio-economic and biophysical connections between the donating and receiving regions. Dry forest ecoregions with high deforestation rates have been particularly underfunded, threatening the ecosystems, species, and local people depending on them. Dedicated global biodiversity fund, raising attention to drier ecosystems, targeting highly threatened areas, and making funding more accessible to local actors for local conservation needs, may help address the gap. This study presented the first subnational level analysis of international conservation funding at the continental scale, revealed the matches and mismatches between the allocated resources and the conservation needs, and shed light on the complex and dynamic landscape of funding opportunities that other actors need to navigate in.

Cette thèse propose une analyse de certaines des politiques considérées actuellement comme étant des éléments-clé de toute stratégie de développement, avec l’objectif de contribuer au récent débat sur le développement international. Je considère en particulier l’élaboration, la mise en oeuvre et l’évaluation des politiques de protection sociale, de développement participatif et d’intégration régionale. Le premier chapitre repose sur l’idée que, pour être efficaces en matière de réduction de la pauvreté, les politiques de protection sociale doivent avoir pour double objectif de permettre aux ménages pauvres d’accéder à des ressources suffisantes pour satisfaire leurs besoins de base, ainsi que de réduire le risque auquel les ménages non pauvres sont confrontés de voir leur niveau de bien-être diminuer sous le seuil de pauvreté. Je propose une méthode permettant d’estimer le degré de vulnérabilité à la pauvreté des ménages. La vulnérabilité est ici définie comme la probabilité pour un ménage de se trouver sous le seuil de pauvreté dans le futur, étant données ses caractéristiques actuelles. Dans le second chapitre, je me place dans un contexte de ciblage des programmes de protection sociale par un score approximant le niveau de vie (proxy-means testing). La précision, et donc l’efficacité, de cette approche pour identifier les ménages pauvres dépendent de la capacité à prédire avec exactitude le niveau de bien-être des ménages, laquelle découle de la sélection de variables pertinentes. Je propose une méthode basée sur l’estimation d’un échantillon aléatoire de modèles de consommation, pour identifier les variables dont la corrélation avec le bien-être des ménages est à la fois élevée et robuste. Ces variables appartiennent à différentes catégories, y compris la possession de biens durables, l’accès aux services d’énergie domestique et d’assainissement, la qualité et le statut d’occupation du logement, et le niveau d’éducation des membres du ménage. Les troisième et quatrième chapitres de cette thèse proposent une analyse ex-post des politiques de développement, et portent en particulier sur les conséquences inattendues d’un programme de développement participatif et les raisons de l’insuffisante performance de politiques d’intégration régionale, respectivement. Le troisième chapitre évalue dans quelle mesure la réaction des deux groupes rebelles présents aux Philippines face à la mise en oeuvre d’un programme participatif d’aide au développement est compatible avec l’idée que ces deux groupes ont différentes idéologies, caractéristiques et raisons pour lutter contre le gouvernement. Il utilise une base de données collectées en utilisant les reportages d’un journal local concernant les épisodes de guerre impliquant ces deux groupes, ainsi que les prédictions d’un modèle d’insurrection basé sur la recherche de rente (rent-seeking). Les résultats sont conformes à la classification proposée de ces deux groupes rebelles ; leur réaction face au projet dépend de leur position idéologique. Le dernier chapitre analyse l’impact des guerres civiles en Afrique sur la performance des communautés économiques régionales, approximée par la synchronisation des cycles économiques des différents partenaires régionaux. Les résultats montrent que la synchronisation des cycles économiques diminue avec l’occurrence de guerres civiles, non seulement pour les pays directement affectés, mais également pour leurs voisins en paix. / In this thesis, I aim to contribute to the recent international development debate, by providing an analysis of some of the policies that are considered key elements of a development strategy. Focusing on social protection, community-based development and regional integration, I consider aspects related to their design, implementation and evaluation. In the first chapter, I propose a method to estimate ex ante vulnerability to poverty, defined as the probability of being poor in the near future given one’s current characteristics. This is based on the premise that effective social protection policies should aim not only to help the poor move out of poverty, but also to protect the vulnerable from falling into it. In the second chapter, I consider the issue of identifying the poor in a context of targeting social protection programs using a Proxy-Means Testing (PMT) approach, which precision, and therefore usefulness relies on the selection of indicators that produce accurate predictions of household welfare. I propose a method based on model random sampling to identify indicators that are robustly and strongly correlated with household welfare, measured by per capita consumption. These indicators span the categories of household private asset holdings, access to basic domestic energy, education level, sanitation and housing. The third and fourth chapters of this thesis provide an ex-post analysis of development policies and focus in particular on the unintended consequences of a community-driven program and on the reasons for the lack of progress in regional economic integration. The third chapter assesses whether the reaction of the two distinct rebel groups that operate in the Philippines to the implementation of a large-scale community-driven development project funded by foreign aid is consistent with the idea that these two groups have different ideologies, characteristics and motives for fighting. It is based on a unique geo-referenced dataset that we collected from local newspaper reports on the occurrence of conflict episodes involving these rebel groups, and on the predictions of a rent-seeking model of insurgency. The findings are consistent with the proposed classification of the rebel groups; the impact of the foreign aid project on each rebel group depends on their ideological stance. In the last chapter, I analyze how civil conflicts affect the economic fate of African regional economic communities through its effect on the synchronicity of regional partners’ economies. I find that conflict decreases business cycle synchronicity when it occurs within a regional economic community, both for the directly affected countries and for their more peaceful regional peers.

Foreign aid has been used on the one hand by donors as an important international relations policy tool and on the other hand by developing countries as a source of funds for development. Since its inception in the 1940s, foreign aid has been one of the most researched topics in development economics. This study adds to this growing aid effectiveness literature, with a particular focus on the under-researched relationship between foreign aid and extreme poverty. The main empirical assessment is based on a sample of 120 developing countries from 1981 to 2013. The study had two main objectives, namely: (i) to estimate the impact of foreign aid on poverty reduction and (ii) to examine the direction of causality between foreign aid and poverty in developing countries. From these two broad objectives, there are six specific objectives, which include to: (i) examine the overall impact of foreign aid (total official development assistance) on extreme poverty, (ii) investigate the impact of different proxies of foreign aid on the three proxies of extreme poverty, (iii) assess whether political freedom (democracy) or economic freedom enhances the effectiveness of foreign aid, (iv) compare the impact of foreign aid on extreme poverty by developing country income groups, and (v) examine the direction of causality between extreme poverty and foreign aid. To achieve these objectives, the study employed two main dynamic panel data econometric estimation methods, namely the systemgeneralised method of moments (SGMM) technique and the panel vector error correction model (VECM) Granger causality framework. While the SGMM was used to assess the impact of foreign aid on extreme poverty, the panel VECM Granger causality was used to examine the direction of causality between foreign aid poverty. The SGMM was used because of its ability to deal with endogeneity by controlling for simultaneity and unobserved heterogeneity, whereas the panel VECM was preferred because the variables were stationary and cointegrated. / Economics / D. Phil. (Economics)