Two Essays on Chinese Loan Market


Student thesis: Doctoral Thesis

View graph of relations


  • Haoyu GAO

Related Research Unit(s)


Awarding Institution
Award date27 Jun 2016


As the largest transitional economy in the world, China has been making persistent efforts towards transforming from central planning banks into market-oriented commercial institutions. Firms in emerging countries like China, rely upon the financing from loan market since they have very restricted access to other external financing channels, e.g. public bond and public equity offerings. This thesis studies two important research questions in banking relevant literature: (1) the characteristics of informationally opaque borrowers and the relationship between borrower information opacity and loan performance; (2) the structure of multiple banking relationships and determinants of concentrated borrowing in China.
The first paper, "Borrower Opacity and Credit Quality: Evidence from China", examines how loan officers assess the financial transparency of borrowers and to what extent this assessment is helpful in predicting credit risk. Specifically, I undertakes a systematic examination of the following two questions: (1) What are the characteristics of informationally opaque borrowers? (2) To what extent does information opacity relate to borrowers’ loan default? The first part of analyses investigates the relationship between traditional information asymmetry proxies and the borrower opacity measure based on loan officers opinions. The second part of analyses builds up the connection between ex-ante borrower opacity and ex-post loan performance.
Using the proprietary data set was obtained from China Banking Regulatory Commission (CBRC), this paper arrives several interesting findings. Different from previous studies assuming a monotonic relationship between opacity and firm size, this paper documents an inverted U-shape relationship. Besides, bank-firm distance can exacerbate opacity problems because it erodes lenders’ ability to collect proprietary intelligence and similarly, geographic dispersion of business groups strengthens member firms’ opacity by increasing the difficulty in gathering information. Consistent with the predictions that screening and monitoring become less effective due to the lack of precise information, this paper finds a significant positive relationship between borrower opacity and loan default risk. The regression analysis shows that, after controlling for a set of commonly used determinants, a one point increase in opacity score raises the likelihood of default by 0.40%. This marginal effect is nontrivial given that the mean likelihood.
The second paper, "Costly Helping: Optimal Structure of Multiple Banking in China", investigates the structure of multiple banking relationships and determinants of concentrated borrowing in China. To begin with, this paper develops a game-theoretical model with complete information on the formation of multiple yet asymmetric borrowing, and test our predictions using credit data from China. The central message of this paper builds upon the stylized facts that five largest state-owned commercial banks, compared with the other twelve commercial banks, are more likely to continue to provide further financing support in event of financial distress but they will be more costly. Based on this basic tradeoff, this paper tries to find theoretical explanation behind the optimal borrowing decisions in terms of the structure of multiple banking.
The game-theoretical model yields three intuitive and testable predictions: (1) firms who need more external funds to recover from potential distress borrow more from large banks; (2) firms who have better opportunities to succeed in the long term borrow more from large banks; and (3) firms whose non-bank financing sources are more costly borrow more from large banks. To verify the hypotheses, this paper also uses this unique Chinese bank loan data for privately-owned listed firms spanning from January 2007 through December 2012. Empirically, this paper yields several pieces of significant evidence. First, large banks continue to extend more credit to borrowers when they become distressed compared with small ones. Second, other things being equal, firms that borrow more loans from large banks incur higher financial expenses. Third, consistent with model expectations, firms with high cash holdings and high profitability rely less on large bank financing and thus tend to borrow more from small joint-stock banks. While, firms with higher growth opportunity and higher cost of non-bank financing sources are likely to build more connections with large banks.
These two essays contribute to the literature in several aspects. The findings in this thesis are among the first to provide evidence on the relationship between the traditional proxies on information asymmetry and borrower’s information opacity based on loan officers’ opinions. Also, the significant positive association between ex-ante opacity and ex-post loan default probability adds to the understanding on the role of soft information in predicting loan default. More importantly, this thesis propose a new explanation for the formation of multiple but asymmetric borrowing, which is especially applicable to developing countries where the banking sector is not fully competitive and banks differ greatly in their helping attitudes for financially distressed firms.