On the Analysis of Firm Value and Idiosyncratic Volatility

This dissertation consists of three chapters covering the following topics in firm value and volatility: valuation of agency cost, valuation of the underpricing in IPOs and the idiosyncratic volatility of public firms. In Chapter 1, I briefly introduce three topics studied in my dissertation. In addition, I summarized the stochastic frontier model which is employed in the study of valuation of agency cost and the underpricing in IPOs. In Chapter 2, I extend the agency cost literature multifold. First, by using the data of the 1,500 S&P Super Composite Index Constituents for 1994-2011, I estimate firm-level agency cost and the uncertainty in firm\u27s maximum benchmark value, respectively, as the mean and variance of the inefficiency term conditional on the composed error. The estimation results reveal that, on average, a sample firm is around 18% and 15% below its optimal value for the period 1994-2006 and 2007-2011 respectively. The variances of the inefficiency term are 0.01 and 0.001 for two periods respectively. Second, using this measure of uncertainty, I construct the confidence interval for the agency cost of each firm. Third, a new concept called Coefficient of Uncertainty of Market Value due to the principal-agent problem (CUMV) is defined and calculated, which measures uncertainty in the benchmark value per unit of agency cost. Finally, I decompose the change in market value of a firm into three components, i.e. change due to agency cost; change due to operational efficiency, and change due to the evaluation of the whole market (called the market effect). I find that the reduction of agency cost and the expansion of the whole market do contribute to firm growth, but the majority of the growth for the sample firms is explained by the improvement of firms\u27 operational efficiency. In chapter 3, I estimate the magnitude of the underpricing of the initial public offering (IPO) for 338 firms during 2001-2010 under the framework of Stochastic Frontier Approach. The magnitude of the underpricing in IPOs and the uncertainty in IPOs\u27 maximum benchmark value are estimated as the mean and variance of the inefficiency term conditional on the composed error respectively. I note that the new issues of a firm with initial offering in US between 2001 and 2010, on average, fall short of 22.9% of their optimal value with a variance of 0.63. As an extension of existing literature, I do not only estimate the frontier model, but investigate the determinants of the underpricing of the IPOs. The estimation results support the fact that the underpricing would be lower if the firms have more reputable underwriters, more insider ownership and higher age at the time of offering new issues. Finally, I introduce a new concept, the Coefficient of Uncertainty of Firm Value due to the underpricing in the IPOs (CUV), which reports the firm value uncertainty for each unit of the underpricing in IPOs. I observe that, on average, the CUV is 4.21 for a sample firm, which implies that firm\u27s uncertainty is indeed sensitive to the underpricing in IPOs. In chapter 4, I investigate the idiosyncratic volatility and its relation to executive ownership during 1992 to 2011. The ownership of executives is employed as the proxy for the behavior of executives to study how executives influence firms\u27 idiosyncratic volatility. Inconsistent with the previous literature, I don\u27t find upward trend of the aggregated idiosyncratic volatility during 1992 to 2011. Instead, I observe that the aggregated idiosyncratic volatility exhibits indeterministic pattern during this period. Moreover, I also note that the reverts of aggregated idiosyncratic volatility occur at a time of the US stock market crash in 2000 and the period of most recent recession (2008-2009). The most interesting finding of this study is that although the idiosyncratic volatility increases in executives\u27 ownership, the idiosyncratic volatility\u27s growth rate is not always positive related with executives\u27 ownership. In Chapter 5, I conclude this dissertation