Governance of timber harvesting and forest conservation in European Russia: Identifying drivers of legal and illegal resource extraction using spatial panel data.

摘要

In this dissertation I use the spatial and temporal variation induced by the collapse of the Soviet Union to estimate the effect of political and economic institutions on timber harvesting and forest conservation in post-Soviet Russia. The dissertation consists of three chapters. In Chapter 1, I identify the drivers of timber harvesting in 32 regions in European Russia by combining remote sensing data on forest disturbance with panel data on determinants of timber supply. I find that timber harvesting followed neoclassical economic theory in post-Soviet Russia, with regional differences in institutional and economic conditions having large and statistically significant impacts on harvesting. In Chapter 2, I estimate the impact of regional governance on timber harvesting. Estimating this impact of governance is complicated by the endogeneity of institutions and correlation between institutions and growth, and I minimize these biases by using within-country data, combining remote sensing data with national statistics, and instrumenting for economic growth. I find a statistically significant and non-monotonic effect of governance on logging: logging increases with a marginal improvement in governance up to a turning point, and then decreases. In Chapter 3, I evaluate the effectiveness of protected areas on forest conservation before, during, and after the collapse of the Soviet Union. I use state-of-the-art impact evaluation techniques, including propensity-score and covariate matching, and combine these with panel data to control for unobservables that would bias the measure of protected area effectiveness. I find that strict protected areas prevented logging only after 1995, and were not effective before or during the collapse of the Soviet Union. Other types of protected areas were not effective at preventing logging within their boundaries during any time period. The main contribution of my dissertation is the finding that spatial and temporal differences in implementing and enforcing institutions -- a result of divergences in governance among Russian regions -- have a direct effect on land use outcomes. Empirically, using spatially-explicit panel data provides a more robust strategy to estimating the drivers of land-use change than previous analyses that relied on cross-sectional samples or national statistics.