Essays on the Economics of Environmental Boundaries

The first chapter investigate the global evolution of economic damages arising from climate-change related natural disasters, showing increasing and highly non-linear dynamics in associated costs. Climate change is arguably one of the most serious challenges that mankind will face in the near future, and it has already been linked to an increase in the frequency and intensity of natural disasters in areas of the globe. Nonetheless, scholars are still debating on whether such increase is mirrored by higher associated economic damages. Here we test the hypothesis that human-nature interactions are characterized by intrinsic non-linearities, in the form of convex and upwardly curved damage functions - linking environmental stressors (e.g. wind speed in hurricanes) to experienced damages. Thus, a shift in environmental stressors distribution would translate into a barely detectable increase in mean damages, but a substantial surge in extreme ones. Applying quantile regression on global data on economic damages from several natural disaster, we show a progressive rightward skewing and tail-fattening of the damage distribution. The second chapter moves to the modeling side proposing a novel Agent-Based Model of agriculture and land use. Building on the evolutionary tradition, we present a spatially explicit model populated by boundedly rational, interacting, heterogeneous farmers. Farmers adaptively hire workers, devote resources to innovation and imitation activity, buy new terrains when in need to, depending on the perceived state of surrounding environment (e.g. market shares or demand level), generating endogenous growth in soil productivity through innovation, imitation and knowledge spillovers. They sell food produced by combining labor and land in a centralized markets, with market feedbacks further influencing their behavior. This complex interaction between agents and surrounding environment allows for non-trivial effects of institutional and behavioral factors on food production, market severity, deforestation, land concentration and seclusion of less productive areas. The model can be employed a useful laboratory to explore scenarios of environmental boundaries. In this chapter we present and explore the main features the model. We show that is it capable of replicating main stylized facts of the agricultural sector, and to generate significant responses to different levels of imitation and knowledge spillovers, market selection and spatial segregation, further testing applications to soil degradation and climate impacts. The third chapter uses the model presented to investigate transition dynamics towards a sustainable regime under scenarios of soil degradation. The progressive net extraction of soil nutrients due to environmentally harmful agricultural techniques, which granted steady increases in soil productivity during past decades, represent a serious threat to future food security. Patterns of stagnating or even descending yields have already been observed in several regions. We model two distinct agricultural regimes: one characterized by high levels of yield in the short run, but progressively eroding soil productivity due to net extraction of nutrients (conventional farming), the other one characterized by lower levels of yield but no soil depletion in the long run (sustainable farming). We here explore how behavioral factor and distinct soil degradation scenarios influence the ability of the system to favor a timely abandonment of harmful conventional techniques, preventing massive deforestation and permanent food scarcity. We document a poor transition capacity in absence of supporting policies. We finally show that unconditional taxes have very little effect on transition likelihood, while effective policies must be aimed at reducing the productivity gap between conventional and sustainable techniques.