1.3 Research Contents, Methods and Framework
1.3.1 Research Contents
Chapter 1 is the introduction of this study. This chapter introduces the practical and theoretical background against which this study is conducted. Moreover, this chapter puts forward research questions and suggests potential research significance based on the practical situations and limitations of prior research. Finally, this chapter briefly states the contents, methods and framework of this research.
Chapter 2 reviews the relevant literature. Centering on the research topic of this study, this chapter reviews existing literature about the dual audit system, the cost of equity and audit quality. It introduces their definitions proposed by previous studies. The main research findings are summarized separately about the consequences of the dual audit system and its change, the determinants of the cost of equity and the determinants of audit quality. Through the literature review, this chapter points out the shortcomings of the existing studies and explores their inspirations for this study.
Chapter 3 introduces the institutional background and conducts the theoretical analysis. First, this chapter provides a detailed introduction of the institutional background—from the establishment to abolition of the mandatory dual audit system for AH companies. Then, it elaborates on the main theories related to this study. Finally, based on the relevant theories and the theoretical analysis of the economic consequences of the deregulation of the mandatory dual audit system on the capital market and audit market, this chapter proposes the theoretical framework of this study.
Chapter 4 investigates the economic consequences of the deregulation of the mandatory dual audit system on the capital market. The objective is to answer the first proposed question of this study. This chapter analyzes the effect of complying with the mandatory dual audit requirement before the deregulation relative to voluntarily choosing the dual audit after the deregulation on the cost of equity. The empirical results in this chapter support the hypotheses about the lower cost of equity for complying with the mandatory dual audit requirement than for voluntarily choosing the dual audit.
Chapter 5 investigates the economic consequences of the deregulation of the mandatory dual audit system on the capital market. The objective is to answer the second proposed question of this study. This chapter analyzes the effect of voluntarily choosing the dual audit after the deregulation relative to using only the single audit in the same period on the cost of equity. The empirical results in this chapter support the hypotheses about the lower cost of equity for voluntarily choosing the dual audit than for using only the single audit.
Chapter 6 investigates the economic consequences of the deregulation of the mandatory dual audit system on the audit market. The objective is to answer the third proposed question of this study. This chapter analyzes the effect of canceling the dual audit relative to keeping the dual audit after the deregulation on audit quality. Furthermore, this chapter deeply explores the role that the loss of a positive spillover effect from Hong Kong-to mainland-based auditors plays in the audit-quality consequences. The empirical results support the hypotheses that keeping the dual audit ensures high audit quality due to the existence of the positive spillover effect from Hong Kong-based auditors, while canceling the dual audit impairs audit quality owing to the loss of this positive spillover. These findings are based on a novel measure constructed in this chapter to represent the spillover effect from Hong Kong-based auditors.
Chapter 7 concludes the study. This chapter presents the main conclusions, policy implications, contributions, and limitations of this study. Based on which, this chapter puts forward potential directions for future research.
1.3.2 Research Methods
This study combinedly uses theoretical analyses and empirical analyses as its research methods. Specifically, based on theoretical analyses, Chapter 2 reviews literature relevant to the dual audit system, the cost of equity and audit quality with a discussion of the shortcomings of the existing studies. Next, through theoretical analyses, Chapter 3 elaborates on the institutional background and related theories, and then constructs the theoretical framework of this study. Su-bsequently, based on the theoretical framework, Chapters 4 to 6 propose the research hypotheses and conduct empirical analyses for hypothesis testing. Finally, Chapter 7 theoretically analyzes and discusses the empirical results, draws main research conclusions, summarizes the contributions, and notes the limitations of this study. The main methods of empirical analyses used in this study are listed below.
Descriptive statistical analysis. This is a basic analysis in the empirical process. In general, the results of this analysis include number of observations, mean value, median value, standard deviation, maximum and minimum for each variable of interest. This analysis helps researchers and readers learn about the general profile and overall distribution of a set of data. In this study, this analysis is used in all chapters involving empirical analyses.
Univariate analysis (mean difference test). This is also a basic analysis in empirical research. It presents the difference (and its significance) in the mean values of each variable of interest between two subgroups of sample. This analysis helps researchers and readers learn about differences in observable characteristics between two subgroups. Moreover, researchers usually can obtain preliminary results consistent or inconsistent with a proposed hypothesis by observing the difference in the mean values of the dependent variable between two subgroups divided based on the values of the independent variable of interest. This analysis is used in all chapters involving empirical analyses in this study.
Multivariate regression analysis[Ordinary Least Squares (OLS) model and Logit model]. This analysis is conducted to examine the relationship of the dependent variable with an independent variable after controlling for the impacts of other independent variables on the dependent variable. In this study, given that all dependent variables in the baseline models shown in Chapters 4 to 6 are continuous variables, the OLS regression analysis is used as the main empirical method for hypothesis testing to examine the relationship of the continuous dependent variable with the independent variable of interest under the condition that the impacts of other factors on the dependent variable are controlled. The Logit model is adopted for cases in which the dependent variable is a dummy variable. In this study, the Logit model is used in the propensity score matching process in Chapter 5 to predict the probability of choosing to use the dual audit for each observation.
Propensity score matching (PSM) analysis. This method is used to deal with potential self-selection bias and impacts of confounding factors. After PSM, the treatment group and control group can usually be tightly matched, facilitating a more accurate and reasonable comparison between them. In this study, the PSM analysis is used in Chapter 5 to rule out the possibility that the difference in the cost of equity between voluntary dual audit and single audit is driven by differences in firm characteristics.
Change analysis. This analysis is used to examine the relationship of a change in the dependent variable with a change in an independent variable after controlling for the impacts of changes in other independent variables. This method can also help address omitted variable issues by differencing out time-invariant factors. In this study, the change analysis is used in Chapter 6 to capture the impact of the loss of the positive spillover effect from Hong Kong-based auditors on the decline in the audit quality of companies canceling the dual audit.
Impact threshold for a confounding variable (ITCV) analysis. This approach is becoming widely used to address endogeneity issues in empirical research by judging to what extent an unobserved confounding variable can turn a statistically significant OLS result into a borderline insignificant one. In this study, the ITCV analysis is used in Chapters 5 and 6 to deal with potential self-selection bias.
1.3.3 Research Framework
Figure 1-3 illustrates the research framework of this study.
Figure 1-3 Research Framework
(1) A shares refer to the common shares issued by a company registered and listed in mainland China. They are traded by domestic institutions, organizations, or individuals with par values indicated in Renminbi (see https://baike.baidu.com/item/A%E8%82%A1/86002?fr=aladdin).H shares refer to the shares issued by a company registered in mainland China and listed in Hong Kong. They are traded by overseas investors with par values indicated in Hong Kong dollar. Institutional investors in mainland China can also directly trade H shares (see https://baike.baidu.com/item/H%E8%82%A1/273519?fr=aladdin#2).
(2) See the news reported by Sina Finance in 2011 and 2012 (http://finance.sina.com.cn/stock/hkstock/hkstocknews/20110418/08099706108.shtml; http://finance.sina.com.cn/stock/hkstock/hkstocknews/20121217/130814026393.shtml).