In my first essay, I investigate the relationship between IPO long-run underperformance (Ritter, 1991) and the idiosyncratic risk puzzle (Ang, Hodrick, Xing and Zhang, 2006), the phenomenon of abnormally low returns for stocks with high idiosyncratic risk. I show that IPO long-run underperformance is in fact a manifestation of the surprisingly low returns for high idiosyncratic risk stocks. IPO underperformance disappears after I control for the idiosyncratic risk. Specifically, the underperformance of IPO firms only presents following the months in which they are classified into the highest idiosyncratic risk quintile. On the other hand, I find that the idiosyncratic risk puzzle is magnified by the IPO underperformance for two reasons. First, IPOs are over-represented in the highest volatility quintile. Second, while stocks in the highest volatility quintile underperform in general, the intra-quintile underperformance is substantially more severe for the IPO firms. My results are robust to different sample requirements. My second essay examines school quality and quality risk capitalization when school quality is uncertain, taking into account uncertainty induced by low signal content in quality measures available to parents or stochastic quality outcomes. Extending the residential bid rent theory to the uncertainty environment, the theory shows that greater school quality increases housing prices steepens the price gradient, whereas the quality risk decreases the housing prices and flattens the price gradient. The empirical models incorporate two sources of quality risk, the variance in measured school quality and school attendance zone instability. Coupling an output based measure using the over-period average of school normalized math test scores based on the Orange County public elementary school average scores with an input based measure using student/teacher ratios provides quality measures that appear to correlate sufficiently with parents' perceptions of elementary school quality, but school peer effects play important role as well. Estimates reveal capitalization of quality and uncertainty that are consistent with theory as well as systematic patterns across housing market phases and neighborhood in income level. My third essay is a meta-analysis of the body of empirical results for school quality capitalization in house prices. One puzzling aspect of the housing markets literature is that, while public school quality is a major concern of many households, empirical studies of school quality capitalization into house prices yield mixed and sometimes inconsistent results not only across studies, but also within studies when using different school quality measures and models. These differences are reflected in the capitalization coefficient value, level of significance, and even direction of capitalization effects. This paper conducts meta-analysis of the school quality capitalization estimates to identify the factors contributing to this variation. It reveals that the way the school quality is measured matters. Peer effects measures yield less significant capitalization estimates than input and output based measures and value added measures exhibit lower significance than other output based measures. Moreover, both boundary fixed effects and neighborhood fixed effect approaches can effectively and significantly control for the influence of neighborhood amenities. Adding more school quality variables reduces the capitalization significance of individual school quality variables. The most unexpected finding is that school quality capitalization significance is much less in the South than in other regions. Also surprising is that econometric methods do not appear to be driving results.

Graduation Date





Chen, Honghui


Doctor of Philosophy (Ph.D.)


College of Business Administration

Degree Program

Business Administration; Finance









Release Date

December 2016

Length of Campus-only Access


Access Status

Doctoral Dissertation (Open Access)