Yushu Zhu

Endogeneity in Accounting and Finance Research: Natural Experiments as a State‐of‐the‐Art Solution

This paper provides a discussion of endogeneity as it relates to finance and accounting research. We discuss the textbook solutions: two-stage least squares, instrumental variables, differenced generalized method of moments (GMM) and system GMM and provide a unifying framework showing how they are related. We consider the limitations of these techniques and then detail a state-of-the-art solution, utilizing a natural experiment as a way of mitigating endogeneity and building stronger theory.

Capital structure: The case of firms issuing debt

This study reinvestigates the relationship between financial leverage and firm characteristics in a cross-sectional setting and a panel setting. Monte-Carlo simulation-based inference results confirm the finding of Barraclough (2007) that a cross-sectional multiple regression model sharing common divisors suffers from a latent spurious ratio problem. To avoid the spurious ratio problem, variables in changes instead of ratios are adopted in two panel models: a first-differenced fixed-effects panel model and a dynamic Generalized Method of Moments panel model. The two models respectively integrate fixed effects (e.g. the persistence nature of financial leverage) and endogeneity features of financial leverage decisions. Model results suggest past realization of debt explains most of the current debt level after controlling for endogeneity. We find no significant association between debt and firm characteristics. JEL Classification: G32, H20

Determinants of long-term debt issuing decisions: An alternative approach

This paper proposes a probit model to test capital structure theories. Our model is designed to circumvent a methodological problem in traditional regression analysis stemming from the use of scaling (e.g., in the form of financial ratios) that has plagued capital structure studies for more than 100 years (Pearson, 1897). Without correction, this problem potentially yields a spurious relation between the dependent and explanatory variables. For example, a negative relationship between capital leverage and firm profitability is concluded in the literature, which is counter to what is observed in reality. Using a sample of leverage increasing cases resulting from public debt issuances in US markets between 1996 and 2006, our probit model results indicate that leverage increasing firms tend to be more profitable. Our finding is consistent with the fact that, in practice, more profitable firms usually have easier access to debt markets.

Political connection, market frictions and financial constraints: evidence from China

This paper studies the relationship between market frictions and political connections in determining financial constraints. We develop a novel index to measure the depth of political connections (PC) at the firm level and provide robust empirical evidence that firms in China actively build PC to alleviate the costs of market frictions. Specifically, we find that firms facing severe market frictions are not as financially constrained as expected. This is because these firms also possess strong PC, which alleviate the costs of market frictions. We find that market frictions can significantly affect financial constraints in Chinese firms, but only for those firms with modest levels of PC.

Pitching DEA Against SFA in the Context of Chinese Domestic Versus Foreign Banks

The primary motivation is to show how the efficient frontier methods data envelopment analysis (DEA) and stochastic frontier analysis (SFA) can be used synergistically. As part of the illustration, we directly compare locally incorporated foreign banks with Chinese domestic banks. Both DEA and SFA reveal that foreign banks are less efficient. DEA shows the main source of inefficiency for foreign banks as managing interest income, whereas domestic banks are inefficient in managing non-interest income and interest expense. SFA reveals contextual variables such as interbank ratio, loan-to-deposit ratio and cost-to-income ratio are significant in explaining inefficiency. The correspondence of rankings based on DEA vs. SFA is positive and moderate in strength but efficiency estimates do not belong to the same distribution. Using DEA and SFA side-by-side can encourage more rigorous and in-depth bank efficiency studies where each method’s limitation can be overcome by the other.

A Timely Comparison of Foreign Banks in China Against Domestic Banks

We employ the efficient frontier methods of data envelopment analysis (DEA) and stochastic frontier analysis (SFA) to estimate efficiency of Chinese bank production in a study that directly compares locally incorporated foreign banks with domestic banks. Both methods reveal that, on average, foreign banks are less efficient than domestic banks, although SFA identifies a stronger contrast between the two cohorts. A break-down of the sources of inefficiency identifies interest income among the foreign banks as an area for major potential improvement, whereas the domestic banks appear to suffer mostly from inefficiencies in managing non-interest income and interest expense. There is a significant positive impact of market share on the efficiency of generating interest income as well as non-interest income. Interbank and cost-to-income ratios have a negative impact on efficient generation of interest income but no significant effect on non-interest income. On the other hand, regulation of the loan-to-deposit ratio does not influence interest income. DEA and SFA efficiency estimates are significantly positively correlated with each other as well as with key financial performance ratios, thus opening the way for both methods to be used together with confidence and in a complementary manner.