This dissertation is composed of three essays. In the first essay, I construct a supply and demand model for crude oil markets. I then fit the model to historical price and quantity data to be able to project future oil prices. Ex-post forecasts using this model predict historical price trends more accurately than most oil forecasting models. The second essay incorporates the supply and demand model from the previous paper into a complex systems model that also includes oil futures markets. Adaptive-agent investors in futures markets choose from a set of rules for predicting future prices that includes the rational expectations equilibrium rule, as well as rules that rely on more short-term information. The set of available rules evolves following a genetic algorithm; agents choose which rules to follow based on their past performance. While outcomes vary depending on the specific assumptions made, under a plausible set of assumptions investors can fail to anticipate shortages properly, leading to significant price spikes that would not occur in the rational expectations equilibrium. The last essay addresses the impacts of carbon cap-and-trade policies on consumers. I calculate how higher carbon prices would affect the prices of different consumer goods, how consumers would respond to the price changes, and how the price changes, along with revenue recycling, would impact consumers of different income levels.

Cette recherche s'intéresse à la modélisation des éco-industries et aux propriétés qui leurs sont généralement attribuées. Précisément, mettre l'accent sur les éco-industries requiert de faire des suppositions dont les conséquences ne sont pas triviales. Ce travail détaille ces conséquences ainsi que la pertinence de certaines. Le premier chapitre considère des éco-industries polluantes, alors que l'analyse économique sur ce sujet assume généralement que les biens et services environnementaux suppriment les polluants. Ce chapitre montre le mécanisme par lequel le marché distribue une activité aux éco-industries en fonction de leur technologie. Le deuxième chapitre enquête sur les limites à supposer une allocation de pollution positive à la solution optimale alors qu'une élimination complète serait possible. L'exemple de l'amiante montre qu'il existe des situations où la pollution-zéro est préférable à un niveau positif. Ce chapitre montre sous quelles conditions ce postulat est vrai. Le troisième travail se concentre sur la notion de services environnementaux. Ce chapitre montre que ceux-ci sont parfaitement substituables aux autorisations de polluer. L'Etat peut donc sélectionner le prix des services environnementaux et il devient possible d'établir une allocation optimale des ressources même en présence de compétition imparfaite. Enfin, le dernier chapitre détaille une supposée fusion entre un pollueur et son fournisseur de biens et services environnementaux. Il montre les répercussions de la fusion sur une firme non-intégrée et amène une discussion sur la différence entre réduction de la pollution en procédé-intégré ou en bout-de-chaîne. / This research focuses on the modeling of ecoindustries and the properties commonly assigned to them. Precisely, highlighting ecoindustries requires the making of some assumptions with nontrivial consequences. This work uncovers these consequences as well as the adequacy of some of them.The first chapter considers polluting ecoindustries, although economic analysis on the subject generally assumes that environmental goods and services suppress the pollutants. This chapter shows mechanisms through which the market distributes economic activity to ecoindustries according to their technology.The second chapter investigates the limits of supposing a positive pollution allocation at the optimal solution although complete suppression is feasible. The example of asbestos shows that there are situations in which zero-pollution is preferred to a positive amount. The chapter shows the conditions under which this hypothesis is true.The third work focuses on the concept of environmental services. This chapter shows that environmental services and pollution allowances are perfect substitutes, then the government can decide on their prices so that it is possible to decentralize a first-best allocation of resources even in the presence of imperfect competition.Finally, the last chapter details a fictional fusion between a polluter and its environmental goods and services supplier. It shows the consequences of such fusion on an unintegrated firm and opens a discussion on the difference between process-integrated and end-of-pipepollution abatement.