Essays on informational intermediation and cost of trading in municipal bond market.
详细信息
文摘
In this dissertation I study the influence of informational intermediary in opaque and illiquid market by exploring the impact of en masse credit rating upgrade,or "recalibration",on yields and cost of trading of municipal bonds. In the first essay,I provide institutional context of the market and outline each rating agencys approach to municipal credit ratings upgrade. In particular,I discuss why credit ratings are important for municipal bond issuers and how most recent theoretical models of credit rating quality can be used to explain why municipal debt had been underrated compared to corporate debt with similar default probabilities and expected recovery rates. Then,I provide a comprehensive analysis of the political and economic forces that prompted credit rating agencies to recalibrate municipal bond ratings in April-May 2010. This recalibration was mechanical i.e.,it was not accompanied by improvements in obligors credit risk profiles) and resulted in a massive upgrade of about 70,000 ratings of municipal bonds. In the second essay,I study the effect of ratings recalibration on the yields of municipal bonds. In contrast to predictions of the semi-strong efficient market hypothesis,I document significant post-recalibration yield declines at the long end of the yield curve for securities with highest upgrades. I explain this reaction by the combination of municipal market segmentation by maturity and individual investors overreliance on credit ratings per se. Municipal bonds are extremely heterogeneous,municipal financial reporting is opaque,so the cost of acquiring and processing information is very large for a non-professional municipal investor. As a result,individual investors rely excessively on credit ratings as a convenient credit risk summary. In the third essay,I examine the effect of ratings recalibration on the cost of trading. I discuss various methods of measuring trading costs and their limitation in the context of municipal bond trading. I estimate the cost of trading using two benchmark methods. I also suggest a third method,a new approach that builds on the econometric technique developed by Harris and Piwowar 2006). The obtained empirical results indicate that trading cost declined for more liquid securities and increased for less liquid securities and retail size trades after recalibration. I also document significant price clustering that deteriorated in the aftermath of ratings recalibration. These results contradict the predictions of market microstructure theory and suggest that some dealers were able to further increase their economic rents after recalibration.