When bank loans are bad news: Evidence from market reactions to loan announcements under the risk of expropriation

Abstract

In this paper we investigate whether inefficient bank loans can reduce the value of borrowing firms when expropriation of the stock of minority shareholders by controlling shareholders is a major concern. Using data from Chinese banks, we find that bank loan announcements generate significantly negative abnormal returns for the borrowing firms. In line with this expropriation view, negative stock price reactions following bank loan announcements are concentrated in firms that are perceived to be more vulnerable to expropriation by controlling shareholders. Finally, we find evidence that a negative relationship between market reactions and firm vulnerability to expropriation exists only when firms borrow from the least efficient banks.

Introduction

Extensive empirical research has shown that bank loans are “special” in the sense that stock markets treat bank loan financing as good news.¹ Financial intermediation theory supports these findings by arguing that banks, as large insiders, can produce information about borrowing firms through initial screening and ex post monitoring (e.g. Fama, 1985, Diamond, 1991). Bank loan announcements therefore convey positive information.²

Nevertheless, this information-production view of bank loans depends crucially on one condition – that lending is efficient. Lenders must have incentive and the ability to screen and monitor borrowers, and lending decisions must be based on the quality of the projects financed, and not on private benefits to bank managers or political considerations.

Although this condition is generally met in US markets, it may perhaps not be met in many other countries, where banks are often controlled by government entities or families with substantial interests in non-financial firms (La Porta et al., 2002, Caprio et al., 2007). This can lead to strong conflicts of interest. Moreover, corruption in the banking sector is a severe problem in many countries (e.g. Beck et al., 2006, Barth et al., 2009). Conflicts of interest and corruption can result in inefficient lending, as shown in recent studies (e.g. La Porta et al., 2003, Khwaja and Mian, 2005).

In an inefficient lending relationship, the informational view of bank loans is doubtful because banks may not screen or monitor borrowers adequately. Furthermore, in an environment in which expropriation of minority shareholders by controlling shareholders is a major concern (La Porta et al., 1999), bank loans can even play an exacerbating role. This is because bank loans increase the amount of resources available to the firm. This, in turn, can increase expropriation of minority shareholders’ interest by majority shareholders, notably through related-party transactions, defaults,³ and payout policies (e.g. La Porta et al., 2000). As a result of expropriation, bank loans can lead to value destruction.

To evaluate these different views on the role of bank loans, we focus in this paper on a single country to provide a better understanding of possible channels through which loan announcements affect stock prices when lending is inefficient. More specifically, we examine stock market reactions to bank loan announcements during the 2001–2006 period in China, using an event-study methodology. The Chinese banking sector has several characteristics that make it particularly suitable for testing our hypothesis. First, Chinese firms experience greater variation in corporate governance structures, which facilitates our identification. Second, the institutional environment in China is weak and the banking sector is inefficient (Allen et al., 2005). In such an environment, expropriation through bank loans can be more severe. Third, in contrast to the US, bank loans are the main funding source for Chinese firms (e.g. Allen et al., 2005). This makes it easier to collect a large, representative sample of loan issuances.

Our analysis of Chinese data shows that bank loans generate significantly negative abnormal returns during announcement periods. Negative stock price reactions following loan announcements are concentrated among firms that are perceived to be more vulnerable to expropriation, including firms engaged in related-party transactions in the years before loan announcements, and those with weak non-controlling shareholders. For borrowing firms that engaged in expropriating related-party transaction activities in the year prior to the loan announcement, we find an abnormal stock price reaction within the next day of −0.898%, compared to firms without any related-party transactions. In addition, we find evidence that the negative association between market reactions and a firm's vulnerability to expropriation is more pronounced when the firm borrows from a low-quality bank. We also find that market reactions are negatively related to the degree of information asymmetry, which contrasts with the information view. Economically speaking, this means that a two-standard deviation increase in stock market volatility (our measure of information asymmetry) reduces the firm's abnormal stock price impact during the first day of trading after a loan has been announced by −1.05% to −1.22%. The findings of this article are not sensitive to alternative event windows, clustering of announcements, self-selection tests, or several other robustness checks. Additionally, our findings are robust to alternative hypotheses, such as political connections.

The findings in this paper are therefore consistent with the expropriation view of bank loans. Because of the significant existence of inefficient lending relationships in connection with political and business ties, corruption, and the prevalence of concentrated ownership in many other countries, our findings may have relevance worldwide.

This study makes several contributions to the literature. First, it contributes to research on relationship banking, and, more specifically, on the value of bank loans. We document new evidence that bank loans can reduce shareholder wealth in an economy with weak institutions. Second, by linking market reactions to bank loan announcements with corporate governance and information asymmetry variables, we find evidence that bank loans may be used as an expropriation tool. Our study therefore extends recent studies on debt expropriation (e.g. Faccio et al., 2005, Bunkanwanicha et al., 2008). Third, this paper responds to the call for more studies on financial and corporate governance systems in which companies have controlling shareholders.⁴

Three papers are most closely related. Harvey et al. (2004) investigate several emerging markets (not including China) and find that announcements of intensely monitored loans which are subsequently internationally syndicated, generate significant positive market reactions, especially among firms that are more vulnerable to expropriation. Their results are therefore consistent with the information view of bank loans. Our study complements their work by using domestic bank loan data from an important emerging market to show the negative impact of private debt financing on firm value. Moreover, instead of using a single proxy (i.e. the separation of cash flow and control rights) as in Harvey et al. (2004) to measure a firm's vulnerability to expropriation, we use multiple proxies. This helps us to better distinguish different views on bank loans. Using US data, Byers et al. (2008) also relate bank loan announcements to corporate governance variables. They find a negative association between market reactions and internal corporate governance quality, presumably due to a substitution effect between banking monitoring and corporate governance. Our paper extends these issues in the context of inefficient lending practices and the risk of expropriation due to weak institutions.

Finally, a recent study by Bailey et al. (2009) examines a similar issue using Chinese data for the period from 1999 to 2004. They also document negative market reactions following loan announcements, and argue that the state ownership of banks is an important explanation for this reaction due to the banks’ conflicting goals of achieving profitability and managing political influence. They do not consider information asymmetry variables. Our study focuses on a test of two competing views of bank loans that are different from Bailey et al. (2009). Our study is therefore complementary to theirs. In addition, our findings are not sensitive to exclusion of the announcements issued by state-owned firms that borrowed from state banks. Our results are also robust to a self-selection test.

The remainder of the paper is organized as follows. Section 2 gives an overview of corporate governance and banking systems in China. Section 3 develops the main hypotheses. Section 4 describes data collection, proxy selection, and the methodology used. Summary statistics for our sample are discussed in Section 5. In Section 6, we summarize the results of our univariate and multivariate analyses for abnormal returns. In Section 7, we provide results for robustness checks. Section 8 concludes.