UC San Diego

Chapter 1 investigates the distributional impacts of a unique car purchase lottery in Beijing on the housing market. I use a difference-in-differences approach to compare heterogeneous neighborhoods before and after implementation of the policy. At an aggregate scale, housing prices within Beijng's fourth ring road increase by 1 to 2% while those outside the fifth ring road decrease by 5%. This is equivalent to transferring about 115,000 RMB ($17,000) from each apartment owner outside the fifth ring to those within the fourth ring. The disaggregate effects are even more pronounced: housing prices increase at locations close to common destinations (employment centers: 5%; primary schools: 3%) and alternative transportation (subway: 3%; buses: 4%). These changes reflect capitalization of the automobile policy and imply a large, and likely unexpected, redistribution across homeowners. The results are relevant to policy, both in the context of unintended consequences and for efforts to develop offsetting measures.

Chapter 2 uses the roll-out of a large transmission expansion in Texas' electricity market to measure the market and non-market impacts of the transmission expansion on benefits of increased renewable capacity. We find large market benefits leading to a payback period of roughly 14 years. However, total welfare improvements from reduced congestion depend on how global non-market externalities are internalized by regional policy makers: accounting for non-market externalities reduces the payback period of this project from 14 to less than 9 years. We discuss the finding's implications for the welfare of regional decisions to build transmission capacity for the U.S. wholesale electricity market in response to federal renewable subsidies.

Chapter 3 investigates the impact of extreme temperature on risk by leveraging daily temperature variation with sales of a "risky" good---Powerball tickets. Temperature and shopping trips are connected through avoidance behavior, leading me to jointly consider trips to retail stores in the analysis. The results show that sales of the risky good per trip increase substantially with temperature, demonstrating a novel dimension in the connection between temperature extremes and human behavior.