Causal inference in spatial environmental economics

Sileci, Lorenzo (2022) Causal inference in spatial environmental economics. PhD thesis, London School of Economics and Political Science.

Text - Submitted Version Download (58MB)

Abstract

This thesis consists in four papers in spatial and environmental economics, in which causal inference methods are employed to analyse two topics: carbon pricing and deforestation. The dissertation is comprised of two parts. The first evaluates the impacts of the 2008 carbon tax implemented in British Columbia, Canada; the second analyses illegal deforestation in Colombia and REDD+ policies in Indonesia. Chapter one studies the effect of British Columbia’s carbon tax on road transportation CO2 emissions and on carbon leakage due to cross-border fuel shopping in the USA. Using the synthetic control method and its extensions, we find that the tax is associated with a decrease in transportation CO2 emissions. However, this effect is not statistically significant, with no role detected for cross-border fuel shopping. In chapter two, we analyse the impact of the tax on PM2.5 concentrations arising from transportation. We detect a statistically significant effect of the carbon tax on air pollution co-benefits, which is heterogeneously distributed across metropolitan areas. Less polluted, less dense and richer areas see greater reductions in air pollution, identifying a post-tax increase in inequality with respect to pollution exposure. Reductions are driven by a switch in commute mode towards low emissions means of transport, principally public transit. Health gains from the tax are large, and regressively co-vary with income. Chapter three focuses on the effects of Colombia’s 2020 Covid-19 lockdown on forest fires. We find that the lockdown is associated with an upsurge in cumulative fires, which is correlated with the presence of armed groups. Chapter four evaluates the effect of the 2011 Indonesian Moratorium on oil palm, timber, and logging concessions. We find that dryland forest inside the Moratorium experienced, at most, a 0.65% rate of forest cover retention compared to non-Moratorium areas, while no effect is detected for carbon-rich peatland. The implied effective carbon price is below US$ 5/tCO2-eq. Moreover, the Moratorium only contributes 3-4% towards Indonesia’s 2015 Paris commitment of a 29% reduction in deforestation by 2030.

Item Type:	Thesis (PhD)
Additional Information:	© 2022 Lorenzo Sileci
Library of Congress subject classification:	G Geography. Anthropology. Recreation > GE Environmental Sciences H Social Sciences > HB Economic Theory
Sets:	Departments > Geography and Environment
Supervisor:	Groom, Ben and Palmer, Charles
URI:	http://etheses.lse.ac.uk/id/eprint/4551

Actions (login required)

Record administration - authorised staff only