| dc.description.abstract | This dissertation contains three essays. The first essay addresses climate impacts on agricultural yields. One practical difficulty in estimating climate impacts is the presence of regionally correlated but omitted factors such as solar radiation and wind speed. Typical panel estimations account for time invariant omitted variables, but do not handle time varying ones that are regionally correlated. To overcome this, an estimation approach incorporating spatial structure is used. We find that the resultant estimates exhibit improved out-of-sample prediction accuracy compared with conventional panel model results but still reveal basic findings found elsewhere in the literature on relationships between temperature and crop yield.
The second essay is about projection of biofuel production and practical considerations involving expensive biorefineries. Many analyses addressing national level expanded biofuel production exhibit unrealistic, time varying locations of facilities. Namely, once built biorefineries are fixed in location, technology and general class of feedstocks they use but these studies ignore such facts. To examine the implications, we do a market penetration analysis with and without that fixity. We find that neglecting asset fixity leads to upwardly biased projections of ethanol attractiveness, as well as unrealistic production variations over time and space. In particular, when asset fixity is considered the price needed to achieve cellulosic market penetration levels comparable to those in legislation is significantly increased, reaching $1.06/liter as opposed to $0.79/liter without it.
The third essay examines renewable electricity and its future market share. Investments in renewable electricity have increased recently due to rapid technological progress. Questions going forward are: (1) Will such technical achievement stimulating market based adoption persist? (2) Are additional developments needed to enhance additional adoption? These questions are addressed in this study using a sector modeling approach. The results indicate that adoption of renewable electricity under current projections of technical progress, will lead to a 25% market share by 2050. If greater market shares are desired, we find this can be stimulated by faster technological progress, reliability enhancing electricity storage and power system management, or direct carbon pricing, with combinations of these supporting as much as a 60% market share by 2050. | en |
