تونل سازی و سرپا نگه داشتن: تجزیه و تحلیل معاملات مربوط به حزب توسط شرکت های فهرست شده چینی

We examine a sample of related party transactions between Chinese publicly listed firms and their controlling shareholders during 2001–2002. Minority shareholders in these firms seem to be subject to expropriation through tunneling but also gain from propping up. On balance, there seems to be more tunneling than propping up. Both types of firms have larger state ownership compared to the rest of the Chinese market but firms that are propped up are larger and have larger state ownership than firms subject to tunneling. Propped up firms are more likely to have foreign shareholders and to be cross-listed abroad compared to firms that are subject to tunneling. Propped up firms also tend to have worse operating performance in the fiscal year preceding the announcement of the related party transaction. Finally, we find that related party transactions representing tunneling are accompanied by significantly less information disclosure compared to related party transactions representing propping.

The vast majority of the academic literature on the expropriation of minority shareholders by controlling shareholders has examined indirect proxies for the likelihood of expropriation.1 There are relatively few papers that examine direct avenues through which expropriation may occur. In this paper, we examine the incidence, characteristics, and valuation consequences of a sample of related party transactions between publicly listed firms and their controlling shareholders in China. We use these transactions to identify the potential for both tunneling and propping up activities by the controlling shareholder. Djankov et al. (2008) note that related party transactions may provide direct opportunities for related parties to extract cash from listed companies through tunneling activities. For example, in April 2001, Tianjin Beacon Paint & Coatings Co. Ltd., a state-owned company controlled by the local government of the city of Tianjin and listed on the stock exchange of Shenzhen, transferred the equivalent of US$4 million in cash to its non-listed state-owned controlling shareholder for safekeeping and management, in a transaction that breached the exchange's listing rules. The market reacted by marking down the stock price of Tianjin Beacon by − 3.4% (after adjusting for general market movements) during the five day window around the announcement of the related party transaction. However, as Friedman et al. (2003) note, related party transactions can also be used to prop up under-performing firms. In a typical example, in August 2001, Luoyang Glass Co. Ltd., another state-owned firm controlled by the local government of the province of Henan and listed on the stock exchange of Shanghai, received a US$28 million loan and bank loan guarantee from its non-listed state-owned parent. Luoyang Glass was an under-performing company whose return on equity (ROE) and market-to-book ratio were below the industry median, while its debt-to-equity ratio was above the industry median. The market reacted by marking up the stock price of Luoyang Glass by 8.1% (after adjusting for general market movements) during the five day window around the announcement of the related party transaction. Finally, related party transactions may also be motivated by purely economic reasons (for example, to re-align the firm's operations). In these cases, the valuation effects of related party transactions would not be expected to be different from the valuation effects of similar arms' length transactions conducted with non-related third parties. In general, prior academic research has focused much more on tunneling than on propping. Identifying the potential for tunneling or propping transactions is an important topic. Though China is emerging as a significant economic power, little research is available on the determinants of performance for firms in a state-controlled economy. Using purchasing power parity exchange rates, the Chinese economy is the second largest in the world, and at current growth rates, it may become the largest in less than 10 years (Allen et al., 2005). The combined market turnover of the two Chinese stock exchanges would make them the second largest in Asia, following Tokyo and ahead of Hong Kong. Increasing numbers of Chinese firms are cross-listed in Hong Kong and in the U.S., and many international investors buy shares in these companies in order to participate in the spectacular growth of the Chinese economy. In our analysis, we examine a sample of 292 related party transactions between Chinese publicly listed firms and their controlling shareholders obtained from filings submitted to stock exchange authorities during 2001–2002. Using filings to the stock exchange as our source of data for related party transactions as opposed to an online dataset such as the China Stock Market and Accounting Research (CSMAR), has both advantages and disadvantages. Reading the filings allows us to observe the entire set of information about the related party transactions that firms make public. It thus allows us to evaluate how extensive the information disclosure is. As we show later on, we document significant differences in the information disclosure between tunneling and propping up related party transactions. We also find significant differences between the information disclosure in the Chinese market and that in the neighboring Hong Kong market. However, since our sample is hand collected, we have to limit the sample period to two years in order to make the task of locating, collecting, and reading hundreds of filings manageable. Time considerations for collecting and processing the filings limit our ability to extend the sample to many years. For this study, we chose our sample period to coincide as much as possible with the Hong Kong data used by Cheung et al. (2006), in order to make comparisons between related party transactions in Hong Kong and China. Nevertheless, as we discuss later, we do not expect that limiting our sample period biases our results. In the first part of our analysis, we examine the characteristics of firms that conduct different types of related party transactions and the characteristics of the transactions undertaken. On balance, there seems to be more tunneling than propping up in our sample. While both types of firms have larger state ownership than the rest of the Chinese market, firms that are being propped up are larger than firms subject to tunneling. Propped up firms are also more likely to have foreign shareholders (B-shares available to foreign investors in the Chinese stock markets), and to be cross-listed abroad (H-shares available to Hong Kong investors, and ADRs available to U.S. investors) compared to firms subject to tunneling. Not surprisingly, propped up firms also tend to have worse operating performance in the fiscal year preceding the announcement of the related party transaction. Finally, we find that value-destroying related party transactions are accompanied by significantly less information disclosure compared to the remaining related party transactions, suggesting that the controlling shareholders who tunnel assets out of publicly listed firms may be manipulating the information disclosure in order to conceal expropriation. In the second part of our analysis, we examine the valuation effects of these related party transactions. We find that the majority of firms in our sample experience a reduction in firm value at the announcement of related party transactions. Our robustness tests show that this reduction in value is not present in similar arms' length transactions, which suggests that related party transactions are unlikely to be motivated by purely economic considerations. We further examine whether proxies for corporate governance, information disclosure, and firm performance are significant in explaining the value destroyed or created in these deals. Firms cross-listed in Hong Kong or in the U.S. earn positive excess returns at the announcement of related party transactions. Firms that voluntarily provide more information about the transactions (firms that voluntarily submit “fairness” opinions from independent financial advisors even when their transactions do not meet the requirements for mandatory submission, for example) also earn positive excess returns. Return on equity (ROE) is negatively related to the excess returns earned by these firms, suggesting that the best performing firms experience the largest value losses in tunneling transactions. Overall, the performance results suggest that controlling shareholders tunnel assets out of firms that have performed well and prop up under-performing firms. Our study contributes to the literature on tunneling and propping in three ways. First, we document the existence of tunneling in a new market, which has only recently attracted the attention of researchers, and we use direct company actions (related party transactions) to do so. This adds to the small but growing list of economies where tunneling has been documented using direct evidence. The Chinese market is also particularly interesting because of the large number of state controlled firms. The previous literature has only examined private controlling shareholders. It is an empirical question whether tunneling also occurs among state controlled firms. We find that it does. Second, although propping has been recognized as the flip side of tunneling (Friedman et al. 2003), there is very little empirical analysis of propping transactions in the literature. By examining propping alongside tunneling in the same market, we can draw comparisons between the two types of transactions and the characteristics of the firms undertaking them. Third, most other studies simply document the incidence of tunneling or propping. We go further than that in highlighting some of the factors — foreign ownership, cross-listing, information disclosure, and firm performance — that affect which firms engage in such transactions and how much is tunneled away or given to the firm as assistance. The paper is organized as follows. Section 2 discusses prior evidence on tunneling and propping. Section 3 describes the institutional background of the Chinese economy. Section 4 describes the data. Section 5 examines the characteristics of the firms undertaking related party transactions. Section 6 examines the valuation effects of these transactions. Section 7 examines which firm characteristics affect the value created or destroyed with the transactions. Finally, Section 8 concludes.

We conclude that minority shareholders in Chinese publicly listed firms that conduct related party transactions with their controlling shareholders are subject to both expropriation through tunneling and gain from propping up from these transactions. On balance, there seems to be more tunneling than propping up in our sample. Both types of firms have larger state ownership compared to the rest of the Chinese market but firms that are being propped up are larger compared to firms that are subject to tunneling. Propped up firms are more likely to have foreign shareholders (B-shares available to foreign investors in the Chinese stock markets), and to be cross-listed abroad (H-shares available to Hong Kong investors, and ADRs available to U.S. investors) compared to firms that are subject to tunneling. Not surprisingly, propped up firms also tend to have worse operating performance in the fiscal year preceding the announcement of the related party transaction. Finally, we find that related party transactions representing tunneling are accompanied by significantly less information disclosure compared to related party transactions representing propping up. Our results are accompanied by two caveats. First, our study is conducted during a period when the Chinese economy has been expanding rapidly. We cannot test whether the Chinese firms which are subject to potential expropriation during periods of growth, are also more likely to be bailed out during periods of economic distress that accompanies economic downturns. Therefore, our results may understate the incidence of propping. Second, we examine large related party transactions, for which stock exchange notification is required. We cannot establish the impact of transactions that are below the thresholds that require stock exchange notification or illegal transactions that companies do not disclose. Our results contribute to the literature in three ways. First, we add China to the small list of countries where tunneling has been examined in studies that use direct proxies for expropriation. Second, we provide a more detailed analysis of propping compared to other studies. As a phenomenon, propping has received considerably less attention compared to tunneling, although it has been argued that essentially they are the opposite sides of the same coin (Friedman et al., 2003). Finally, we highlight the impact of corporate governance, information disclosure, and firm performance proxies both on the likelihood of tunneling or propping and on the value gains or losses that publicly listed firms experience in these transactions. Our results have two regulatory implications for emerging markets. First, we highlight the importance of foreign investors and cross-listing in mitigating the expropriation of minority shareholders. This suggests that the sophistication of the firm's shareholder base may be very significant in preventing expropriation. It also suggests that higher listing and accounting standards provide additional protection from expropriation. These have important implications for allowing foreign investors to participate in emerging markets and in strengthening accounting standards in these markets. Second, our results suggest that higher standards of disclosure with respect to the transaction itself also help mitigate the expropriation of minority shareholders through related party transactions.