RESOURCE MANAGEMENT IN THE AGE OF CLIMATE CHANGE: SUSTAINABILITY AND RESILIENCY
Duah, Isaac Amponsah
MetadataShow full item record
In the first chapter, we show that, after the announcement of a new environmental policy, firms can respond by increasing or decreasing their appropriation and pollution. In addition, we demonstrate that reductions in exploitation levels are more significant when: (1) few firms compete for the resource; (2) firms are not price takers; (3) firms do not impose significant cost externalities on each other; and (4) the resource is abundant. Our results, therefore, indicate that policy announcements can trigger a reduction in resource exploitation before the law comes into effect, helping rationalize empirical observations. In the second chapter, we present an empirical evidence of the green paradox through implementation lag of the Clean Power Plan (CPP). We use a hedonic pricing model and difference-in-difference estimation technique to estimate the reactions of coal resource owners and coal power plants between the announcement and the implementation of the CPP. I find that the announcement of the CPP caused a decrease in price of coal by ¢3.19 per MBtu. This in turn caused an increase in the coal burned by power plants by 0.01%. As further evidence of the green paradox, a price premium for the ash content of coal was reversed after the announcement of the CPP; resulting in coal power plants burning coal with 0.3 percentage points more ash content after the announcement of the CPP. Power outages cost American consumers an estimated $79 billion annually and expected to increase due to the increasing impacts of climate change. Accurate early prediction of weather- related interruptions provides time for power companies to make adjustments to minimize both the geographic extent and duration of realized outages. In the third chapter, we model weather- related power outages using weather forecast information at 24-, 21-, and 18-hour forecasts in Austin, San Antonio, Houston and Dallas. We use a multi-equation estimation approach in conjunction with the Heckman sample-selection model. Our model has up to 80% prediction accuracy and up to 78% true positive and false positive trade-off reliability in the 24-hour forecast.