Three essays on corporate bond markets
China's bond market, which has the second-highest market value of any bond market in the world, plays an important part in China's financial system and national development strategy. This thesis consists of three essays in which I examine Chinese corporate bond market-specific topics.
In the first essay, I investigate the air pollution premium of corporate bonds in China. I find that bonds issued by firms located in more polluted areas are associated with higher offering yields at issue in the primary market and higher yield spreads in the secondary market. Both polluting and non-polluting industries are affected by the air pollution premium. The statistically and economically significant air pollution premium is robust to a battery of sensitivity checks. Furthermore, the results of the difference-in-difference (DID) test show that the air pollution premium is associated with rising investor attention to environmental risks. The air pollution premium is higher among firms with an opaque information environment. Poor air quality does not lead to a decrease in market trading volume or liquidity, suggesting that trading frictions are not the cause of the air pollution premium in the bond market. I also find that air pollution likely does not increase the operating costs and default risks of firms, nor does it reduce firms’ credit ratings.
In the second essay, I examine the behavior of major shareholders in convertible bond issuance. The subscription structure of China’s convertible bond market does not follow the pattern found in the traditional convertible bond financing theory. I investigate whether firms signal outside investors via the subscription ratio of major shareholders when issuing convertible bonds. I find that a high shareholder subscription ratio is often a misleading signal fabricated by low-quality firms to attract external investors, resulting in lower lot winning rates of convertible bonds. These firms have relatively poor post-issuance performance and experience a significantly negative market reaction within one year of convertible bond issuance. This finding is robust to alternative performance measures and the DID test of the revised subscription policy. I find that this effect is more pronounced in firms with limited political connections, with more information asymmetry, and with lower standards of corporate governance.
In the third essay, I investigate the market pricing errors in the convertible bond market. I employ the least squares Monte Carlo simulation method (LSM) and build an optimized valuation model to estimate the theoretical value of convertible bonds. I find that the data distribution of convertible bond market pricing errors in China is left-skewed, and the market prices are generally overvalued. I analyze the causes of convertible bond market errors in terms of investor structure and trading system. There are four key findings. First, I find that retail investor sentiment reduces market efficiency. Second, institutional investors behave relatively rationally and are less susceptible to market noise. Third, major shareholders can reduce undervaluation, but their inhibiting effect on overvaluation is relatively weak. Fourth, due to the shortcomings of the T+1 trading system of stocks, investors prefer convertible bonds whose underlying stocks are eligible for short selling to achieve the actual trading effect of T+0. This in turn strengthens the speculative motive of these convertible bond investors, generating greater market price errors.
