Financial Distress without Bankruptcy: The Case of China 1

Transcription

1 Financial Distress without Bankruptcy: The Case of China 1 Joseph P.H. Fan Chinese University of Hong Kong Jun Huang Shanghai University of Finance and Economics Ning Zhu University of California, Davis June, 2007 Preliminary and Incomplete Do not cite without authors agreement 1 Fan is with Chinese University of Hong Kong and can be reached at pjfan@cuhk.edu.hk by or by phone. Huang is with Shanghai University of Finance & Economics and can be reached at sufehuang@gmail.com by or by phone; Zhu is with University of California, Davis and can be reached at nzhu@ucdavis.edu by and by phone. Part of the work was completed when Huang was visiting Chinese University of Hong Kong and Zhu was visiting Beijing University. Zhu acknowledges financial support from Beijing University. All remaining errors are ours only. 1

2 Abstract We study distressed firms in China, where bankruptcy law is weak, to understand what mechanisms influence distressed firm behavior in emerging markets and find that institutional background matters considerably. Ownership structure, government quality, and local financial development are important to operating performance, capital structure, investment policy, and ultimate recovery likelihood. SOEs, which experience loose monitoring and receive more soft lending, are more sensitive to the above forces in distress than private firms do. Our findings provide novel evidence on how institutional factors discipline distressed firm behavior in emerging markets. 2

3 Scholars have devoted numerous studies to understanding bankruptcy, a primary outlet for distressed firms in developed markets (Baird, 2007, Bris et a. 2005, 2006, Khal 2002, Stromberg, 2000, Wruck 1990, among many others). It is widely accepted that bankruptcy laws and close external monitoring by stakeholders (i.e. corporate bond holders and large institutions) play important roles in disciplining financially distressed firms and influencing corporate financial decisions. In contrast, it remains an open question as to how firms deal with distress in markets where bankruptcy law and creditors have less disciplinary power. As a matter of fact, it is fairly common in emerging markets that banks make soft loans (La Porta et al. 2000b, Dinc 2005). At the same time, creditors often have difficulty in liquidating distressed firms or seizing distressed firm assets given the lack of bankruptcy law and its enforcement. Although there is still controversy on whether lending relationship with banks help or hurt distressed firms (Degryse and Ongena, 2005, Petersen and Rajan 1994), it becomes well established by now that distressed firms in emerging markets, because of the weak legal enforcement, have greater bargaining power than their counterparts in the developed markets. The interesting question that we attempt to answer in this paper is how firms react to financial distress given such power? Given the extensive games played between different stakeholders and dynamic operating and financing decisions in financial distress, we find it particularly interesting and important to study how institutional forces shape the behavior of distressed companies in such markets. As a leading emerging economy, China witnesses the common practice of soft lending among its state-owned banks (Allen et al. 2005a). Also like many other emerging markets, interest rates and capital allocation are often distorted by the government and regulations during the fast economic transitions (Maskin and Xu 2001). Due to the lack of bankruptcy law enforcement, distressed firms in China usually do not have to face the threat of liquidation. Instead, firms usually deal with distress by adapting their corporate policies or renegotiating lending contracts. We take advantage of this opportunity to understand how institutional factors, in particular ownership structure, government 3

4 quality, and financial development, influence distressed firms behavior in China, and in the more general context of emerging markets. Our study draws its data from the comprehensive Annual Industrial Survey Database ( ) by the Chinese National Bureau of Statistics (NBS), which surveys financial and operating information on all industrial state-owned enterprises (SOEs) and non-soes firms (including domestic private firms, joint-ventures, and foreign firms) with annual sales of at least RMB 5 million in China. Such a sample of on average 200,000 companies covers all publicly traded manufacturing companies in China and the dominant share of the industrial sector of the national economy. Our findings can be summarized as follows. The ownership structure of the firm matters considerably to firm behavior in financial distress. State owned enterprises (SOEs) are much slower than private firms in adjusting their policies to deal with their financial distress. SOEs in financial distress display significantly inferior performance (measured in return on sales, total-factor productivity, and earnings growth) to their private counterparts. In contrast, distressed SOEs are more aggressive with their post-distress corporate financing policies, reflected by relatively higher leverage, higher fraction of long-term liabilities, and higher level of external investment. As a result, SOEs on average take longer to emerge from financial distress and are less likely to ultimately emerge from financial distress. Our additional change-in-change analyses further confirm that institutional factors not only explain the cross-sectional differences in average firm behavior, but also are particularly powerful in explaining how firms adapt their decisions in distress. We take a further step to understand how a change in state ownership can influence firm performance and distress outcome. Our evidence suggests that the nontransferability of the state ownership and assets has much to do with SOE s disappointing performance in distress. An increase in private ownership in place of state ownership during distress, even if not big enough to change the dominant state ownership, helps discipline firm behavior. 4

5 Other institutional background also influences how firms deal with their distress. Firms from regions with better quality governments and more developed financial sectors are more responsive to constraints posed by distress. Such firms typically perform relatively better through financial distress and engage in more conservative corporate financing decisions. Our findings are particularly strong for SOE firms, which normally face less monitoring and discipline than private firms do. We conduct a host of robustness tests by implementing different ways of defining distress (the Altman Z-value, leverage, and interest coverage), different criteria for distress (alternative cut-off z-values and separate cut-off values for private and public companies), different specifications (pooled ordinary least squared (OLS) regressions, panel regressions, and change-of-change regressions), and our above main results remain robust. We perform additional robustness tests that explicitly control for the fact that SOEs are more likely to enter distress in the first place, and obtain similar results Our further analyses on the relationship between change in firm performance and financing decisions from pre- to post-distress period confirm that the differences in distressed firms driven by corporate behavioral change within distress is directly linked to the institutional background, as opposed to the alternative hypothesis that our findings are merely continuation of different firm behavior before distress. The above analyses provide additional support that institutional background is critical to the dynamics in distress. This study contributes to the literature in two important aspects. First, we are among the first to focus on how distressed firms shift their behavior, in response to institutional factors. Because China is a large and decentralized country where local institutional features vary widely while the country s managers and firms share similar culture and language background, our single-country study mitigates potential endogeneity issues common in cross-country studies. The paper builds upon the growing literature on the interaction between institutional environment (i.e. law, government, and financial development) and financial markets (for example, La Porta et al. 1998; La Porta Lopez-de-Silanes and Shleifer, 2002). Now that we have come to understand how law and political landscape influence average firm behavior, it is natural to take a step further and investigate how such institutional background manifests themselves through 5

6 important corporate events such as distress. Distressed firms provide some unique opportunities to study institutional background s impacts given their high stakes to various stakeholders and the dynamic changes happening around distress. Although better institutional environment is generally beneficial to the development of financial markets (La Porta et al. 1998, La Porta Lopez-de-Silanes and Shleifer, 2002), the problem becomes more complex for distressed firms and the empirical findings from developed markets tend to be mixed (Hoshi et al and Ongena et al Degryse and Ongena, 2005, Petersen and Rajan 1994). Better institutional background can exert greater pressure on distressed firms and discipline their actions, which better protect the interest of creditors. On the other hand, poorer protection of investor rights may provide distressed firms more flexibility, facilitate recovery from financial distress, and ensure the welfare of share holders and employees. This study provides robust evidence of how government incentives and interventions, a generally important institutional factor, affect firm distress resolutions. Second, the paper extends the knowledge on distressed-firm behaviors in emerging markets whose institutional environments are different from those of developed economies. Consistent with theory s prediction on the disciplinary role of financial distress (Jensen 1986), we find that firms adjust their financial policy in distress (i.e. leverage, investment) in order to recover. In addition to existing findings that firm factors such as capital structure (Booth et al. 2001, Ofek 1993) matter to the speed at which firms respond to distress, we show that institutional factors, such as firm ownership structure, government quality, and financial development, determine firms sensitiveness to distress and how they adapt their decisions in distress. Our in-depth observation of distressed firms in China provides further understanding about out-of-court distressed firm behaviors and its implication on ex-ante firm investment and financing decisions in emerging markets. The rest of the paper proceeds as follows: Section 2 reviews the practice of bankruptcy in China facing distressed firms and develops hypotheses; Section 3 outlines 6

7 the data and empirical methodology; Section 4 presents our findings and discusses the results before we conclude in Section Corporate distress and bankruptcy in China and hypothesis development 2.1. Background of corporate bankruptcy in China After its economic reform and openness started in the early 1980s, China enacted its first bankruptcy law in Given that the law was enacted during the early stage of the economic reform, the law lags the practice in many aspects (China Year Book of Law, 2001). It was established while the state economic sector still dominated the national economy. 2 Many parts of the law intended to resolve problems between state-owned entities, SOEs and their state-owned lenders, and the law focused more on promoting the state economy than resolving conflicts in civil relationship. In addition, given the historical context, the law tended to put SOEs at a relatively advantageous position compared to collective (communal) firms and private companies. Consequently, SOEs were much less bound by the law and were given greater flexibility in distress than other companies were (Allen et al. 2005a). Second, the law does not promote liquidation as a means to deal with distress. Instead, it tends to encourage negotiation and workouts between/among the involved parties. This usually gives SOE managers and other corporate executives greater bargaining power during distress and greater freedom in their management decisions before distress (Du and Xu 2007). Given that many debtors have government backings and share the common belief that they are unlikely to seize their creditors assets or engage in aggressive debt-collecting services, distressed firm executives do not see liquidation as a tangible threat to their operations. Third is weak enforcement. Many details of the law were intentionally missing, which pose challenges to the court system s enforcement. Judges and attorneys alike often find themselves lack the specific clauses to help them make decisions. Instead, the 2 A Completely New Bankruptcy Law in Chinese History, by Hua Song, Aug 30, 2006, Shenzhen Business Journal. 7

8 court system has been very conservative with deciding whether to handle bankruptcy cases and reluctant to grant debt discharge or reorganization, to avoid potential problems arising from inconsistency with the law (Allen et al. 2005b). It is fairly common that the court requires distressed firms to first obtain consent to their bankruptcy decisions from the local government and to propose a satisfactory plan to place its existing employees, before the court even considers handling the cases. As a result, only a small fraction of filed bankruptcy cases were eventually handled by the court system and even less were eventually discharged. Appendix I reports statistics of filed and discharged bankruptcy cases in China. On average, only about 5 percent of all bankruptcy petitions were handled by the court. During 2001, only 315 out of 7,233 bankruptcy cases were handled by the court, with other cases being dismissed or withdrawn (Law Year Book of China, ). As Allen et al. (2005a) point out, such a weak legal environment of the bankruptcy results in weak external corporate governance and imposes little threat and penalty to corporate executives. Corporate executives, particularly those of SOEs, face much less pressure than their foreign counterparts to satisfy the liquidity requirement posed by interest and principal payment. Further, the lack of liquidation as an ultimate distress outlet helps keep executives entrenched and continue their poor decisions without being held accountable in a timely fashion. Finally, because distressed firms are less concerned with their survival prospect, they are slow in adjusting their operating strategy and financing decisions. Instead, many may opt to continue their inefficient investment or even start new investment projects which add additional risks to the firm cash flow. Hence, we believe that such a unique background in Chinese bankruptcy law provides us an opportunity to investigate how alternative governance mechanisms other than the law or its implementation helps discipline executive behavior and improve corporate decision making. It is worth pointing out that it is naïve to believe that because the state government owns both state-owned banks and state-owned enterprises, there is little tension between the creditors and debtors. After extensive reform in the banking sector, 8

9 state banks that used to dominate deposits and lending find themselves facing fierce competition from new and more independent commercial banks. Reduced intervention from the state government make the state banks more discreet with their loans and more watchful should the debtors default. In addition, the recent securitization of state-owned banks requires such banks to be more vigilant with their new lending and outstanding loan recovery. Consequently, these banks are now putting greater-than-ever effort to ensure distressed companies come up satisfactory plans with their defaulted debts. In contrast, distressed company managers and local government officials usually have to face a much wider range of issues from local economic development and employment to retired employees pension financing when dealing with distresses. In addition, executives at distressed-firms have additional incentives to steer companies out of financial distress. First of all, existing literature has documented that financial distress can lead to considerable amount of direct and indirect costs to firms (i.e. Altman 1968, Opler and Titman, 1994). Firms may suffer in the areas ranging from un-trusting suppliers and retailers, difficulty in obtaining credit, to tarnished reputation among customers. Second, many SOE executives are also government officials. Their performance as executives is evaluated by their supervisors and has important implications to their future career promotion (Bai et al. 2000, 2002). In addition, it is not uncommon for state bank officials exert their pressures on SOE executives supervisors, for favorable terms in loan contract renegotiations. Sometimes, the very same relationship that spawns soft-lending in the first place may, ex pose, serve as tools to facilitate debt repayment. Finally, under Chinese securities law, distressed firms are refrained from issuing equity in the near future and listed distressed firms are subject to regulatory measures ( special treatment (ST) and penalty treatment (PT) and even more severe consequence of de-listing. The above discussion suggests cross-sectional variations in distressed firm behaviors due to varying ownership structures, government incentives, and financial sector development. Next, we discuss how these institutional factors can influences the dynamics of financial distress, subject to the weak bankruptcy law. 9

10 2.2. Institutional factors influencing distressed firm behavior A. Ownership structure and financial distress One distinct feature in the economy of many emerging markets is the special role played by SOEs (Bai et al. 2000, Cull and Xu 2003). SOEs often establish their competitive edge through monopolistic market power and favorable tax treatment (Dinc 2005, Johnson and Mitton 2003, Shleifer 1998, Shleifer and Vishny 1994). In addition, extant studies show that SOEs are also helped by favorable financing arrangements: they are much more likely to obtain bank loans than their private competitors and less subject to liquidity constraints posed by lending agreements, often times helped by state-owned banks (Sapienza 2004 Khawaja and Mian, 2006). Like many other emerging markets, such practice is fairly common in China. Given such problems of soft lending and loose monitoring, SOEs tend to be less conservative in planning their investment and corporate financing strategies, and to be more likely to end in financial distress, ex post. Therefore, we later perform empirical analyses that explicitly control for the fact that SOEs are more likely to enter distress and obtain consistent results with our main findings. The ex-post consequence of financial distress puts quite different pressures on managers at SOEs, compared to those at private companies. Managers at private companies, often times also major shareholders, watch carefully the creditors move and have to quickly adjust corporate strategies so as to raise enough cash flow to satisfy interest payment and creditor claims. Although banks rarely liquidate distressed SOE assets or seize their assets, it occasionally happens that banks indeed force auction-offs of non-soe companies. Further, the possibility of not being able to obtain future lending from the existing lenders alone is threatening enough to many private firms to modify their business behavior. In contrast, SOE executives understand that many of the loans that they receive are policy-motivated and banks, especially state-owned ones, are unlikely to terminate the lending relationship or engage in further punitive actions. Consequently, SOE executives are usually slow in responding to financial distress. In addition to the soft lending practice, SOEs distinguish from private companies in their weak governance mechanisms and distorted managerial incentives. Although 10

11 many of Chinese SOEs have gone through securitization and marketization reforms, most SOE executives and board directors are still under government administration. The pay scale of SOE managers is closer to that of the civil servants than that of executives in private sectors. There was little practice of option granting, Employee Share Ownership Plans (ESOPs), or other forms of pay-for-performance. Such a lack of incentive leads SOE managers to be not as motivated as managers at private companies and to care less about firm performance. In addition, such executives at SOEs are often evaluated on dimensions other than pure firm performance, such as contribution to local economic growth and employment, which distracts them from focusing on maximizing the shareholders wealth. On the other hand, executives at SOEs often enjoy greater power and are subject to little supervision from the board. Internal monitoring (i.e. board committees or independent directors) and external monitoring (i.e. institutional investors) are both weak and can rarely effectively limit the executives power. As a result, we expect private firms, which are governed by the more marketoriented mechanism, to respond more quickly to financial distress (reflected by relatively better operating performance), and to more effectively adjust their strategies to repay debts (reflected by more conservative capital structure and responsible financing decisions) and to be more likely to emerge from bankruptcy. In addition, we perform additional study to understand whether both features of the SOEs, namely the access to soft lending and the different managerial monitoring and incentive, are responsible for the different decisions that SOEs make in distress. B. Governmental/legal effectiveness and financial distress Existing studies have shown that institutions and legal environment are critical to economy and financial markets at the macro-level (La Porta et al. 1998) and to corporate decisions and the dynamics between different stake holders at the firm level (Demirguc- Kunt and Maksimovic 1996, 1998, 1999). Better government quality or legal enforcement can provide better protection of lending contracts and enhance credit availability to companies (Charumilind et al. 2007, Durnev et al. 2004, La Porta et al. 1997). In addition, strict legal enforcement of lending contracts may pose greater liquidation threat to companies in distress and urge firms to adjust their behavior more 11

12 promptly. Further, more developed governments are more experienced with dealing with foreign companies. The greater presence of such companies increases the chance of local business obtaining financing through foreign direct investment (FDI). The exposure to sophisticated financing tools and contracts offered by foreign investors is beneficial to local executives to take lending relationship more seriously. Finally, governments and legal effectiveness can considerably modify lender (mostly banks) behavior and indirectly influence firm decisions. In areas where governments are less geared towards market economy, government officials are more likely to value non-economy related activities, such as ideology. Under such influences, government officials are more likely to exert pressure on banks to make policy-oriented loans and interfere with debt renegotiation in the event of financial distress, causing banks to lose their independence in lending out loans. If firms, especially SOEs, believe that government officials will help them with debt negotiation or forgiveness, then managers will lack the incentives of adjusting firm strategies to recover from distress. Hence, the quality of the government influences the expectation that people have regarding the enforcement of the law and contracts, and can influence credit availability and interest rate, ex-ante (Bris and al. 2006, Ofek 1993). In sum, we expect firms from areas with better governments to respond more quickly to financial distress (reflected by relatively better operating performance), and to more effectively reduce their borrowing (Rose-Ackerman 1991) and to be more likely to emerge from distress. C. Financial development and financial distress Extant literature (Cull and Xu 2005) documents that financial development aligns firm actions with the market mechanism. In particular, Maskin and Xu (2001) and Sapienza (2004) show that the banking sector plays an important role in allocating resources and exerting the necessary monitoring on debtors. The emergence of some developed independent commercial banks, an important indication of the financial development, is important so that financial institutions are less likely to be subject to political power and make policy-driven loans. In addition, such commercial banks 12

13 usually stick closely to their lending guide in order to manage risk and avoid nonperforming loan. This screening process enables non-state-owned banks to make safer loans and ensure that they are less likely to get involved in firm distress. Compared to state-owned banks funded by state funds, independent commercial banks are more careful with their lending exposure after they make the loans. If they notice debtors in distress, non-state-owned banks will exert considerable pressure on debtors to repay their loans or engage in active workout to restructure their lending. On the other hand, leading state-owned banks are less prudent with making loans and less effective in recovering their due debts, reflected the high level of non-performing loan for such banks during an extended period of time (Brandt and Zhu, 2000). In sum, we expect better financial development to represent better institutional background to firms in distress. From the firms perspective, managers and firms located in more developed financial markets value contracts more, both from their respect for the law and from their fear of harsher punishment by creditors. Consequently, they try harder to turn distressed company around. Further, a firm from an area with more developed financial sector is less likely to be protected by lax lending practices or policy-driven lending and is hence motivated to make quicker changes to improve firm performance and repayment capacity. Finally, because firms are more likely to deal with market-oriented non-state banks in more developed areas, they care more about their reputation as borrowers to improve their chance of obtaining credit in the future. We expect these companies to take more actions in distress to turn around. As a result, we have similar expectation to the above two aspects of institutional background in that firms from more financially developed areas would respond more quickly to financial distress, more effectively improve their financial health, and recover sooner from distress. 13

14 Summarizing our above discussions on how state ownership, government quality, and financial development can influence distressed firm behavior, we formulate the following specific hypotheses: Hypothesis 1. Better institutional background (lower state ownership, better government quality, and greater financial development) helps improve distressed firms operating performance. Hypothesis 2: Better institutional background motivates firms to be more responsive and steers distressed firms toward more responsible (conservative) capital structure. Hypothesis 3: Better institutional background leads distressed firms to more conservative investment projects. Partly reflecting the consequence of the above three hypotheses, we expect that firms with better institutional background to have a greater chance of recovering financial distress and if they do, recover from financial distress sooner than firms with less favorable institutional background. We formally formulate it in the following hypothesis: Hypothesis 4: Better institutional background helps distressed firms emerge from distress. 3. Data description and empirical methodology 3.1. Data description This study uses the Annual Industrial Survey Database ( ) from the Chinese National Bureau of Statistics (NBS). The NBS surveys financial and operating information on industrial firms and publishes such information in the official China Statistics Yearbooks. Previous studies have widely used such a data source and confirm that the data are accurate and well representative of the national economy (Chow 1993, Chuang and Hsu 2004, and Li et al. 2006). The database covers: 1) all state-owned enterprises (SOEs) regardless of their annual sales, and 2) all non-soes firms (including 14

15 domestic private firms, joint-ventures, and foreign firms) with annual sales of at least RMB 5 million (roughly, $600,000 US, according to the exchange rate on Dec 31 st, 2005). The sample encompasses all provinces in mainland China and the number of firms included in this database ranges from 162,033 to 276,474 across the sample years. All publicly-listed industrial (about 700) firms are included in the database. We adopt several measures to identify firms in financial distress. Our primary criterion to identify financially-distressed firms is the widely-used Z-score (Altman 1968). We focus on firms that are potentially in financial distress within the first three years ( ) of our sample, which enable us time to track the firm dynamics within financial distress between 2001 and Following the methodology in Altman (1968), we estimate the Z-score of each firm and classify firms with z-score of 1.23 and less to be in financial distress (Please refer to Appendix II for details on how we estimate the Z-value and set the cut-off value). A firm has to be classified as distressed for at least two consecutive years (1998 and 1999, 1999 and 2000, or 1998 through 2000) to be qualified as a distressed firm. As a result, we have a total of 18,850 firms included in the distressed firm sample. In addition, we include firms that are financially healthy as the control sample. We require a firm to be out of financial distress during the entire sample years ( ) to be included in the control sample. Such classification generates a total of 17,872 firms in the control sample and provides a similar number of firms to the distressed firms. Note that the number of distressed and that of financially-healthy firms do not add up to the total number of firms in the survey, by such a classification. We experiment with alternative selection criteria for control firms, such as requiring the firms staying out of financial distress between 1998 and 2000 only, and obtain similar results. In addition to the Z-score measure, we use two other criteria: leverage and interest coverage. With leverage, we consider a firm to be in financial distress if the leverage of a firm is greater than one, that is, if the firm has more total outstanding debts than its total assets. With the interest coverage measure, we classify a firm to be in financial distress if its interest coverage ratio is less than one (suggesting that a firm s operating expenses are 15

16 not enough to cover its interest payment obligations). The two alternative classifications generate consistent results to our main results Empirical methodology Based on the discussions in Section 2, we focus in this paper on three aspects of institutional background important to distressed firms behavior: ownership structure, governmental/legal effectiveness, and financial development. We measure state ownership with the following three variables. We first calculate a continuous variable of the fraction of a firm s equity owned by various levels of government agencies. In addition, we construct a dummy variable that equals to one if various levels of government agencies own at least 51 percent of the firm and zero otherwise. Finally, we introduce a categorical variable representing the level of administrative hierarchy of a firm: 1 for central government, 2 for province, 3 for city, and 4 for county or village. Because firms under direct management by higher levels of administrative hierarchy are more likely to receive favorable treatments from different sources and subject to more government intervention, we expect them to be less careful with their decisions in financial distress. The last proxy generates similar results to the previous two and therefore we focus on presenting evidence on the first two variables in the paper. We focus on two particular aspects of the local institutional background: local government effectiveness and development of the local financial markets. We adopt three measures for government effectiveness. The first proxy is the quality of government index obtained from The Annual Report of Urban Competitiveness in China, published by Social Science Academic Press (2003), which measures the level of bureaucratization, the frequency of government expropriation and the level of citizen satisfaction. In addition, we also adopt two alternative measures in Fan and Wang (2002), on the degree of local market openness and the degree of marketization. These two measures generate qualitatively the same results and we focus reporting results on the former to conserve 3 Such results are available from the authors upon request. 16

17 space. To measure how institutional factors influence distressed firm behavior in particular, we also generate interaction dummy variables between the above institutional variables and the dummy variable for distressed firms. Regarding local financial development, we use the following two provincial variables. First, we use the financial development index from Fan and Wang (2002), a widely used measure which captures the competitiveness of banking sector and marketization of credit allocation. The competitiveness of financial industry is defined as the deposits held by non-state-owned financial institutions scaled by regional total deposits. The marketization of credit allocation is the short-term loans to non-state sector (such as agricultural loans, loans to village/township enterprises, loans to private enterprises, and loans to foreign-owned enterprises) divided by total short-term loans in a region. Second, we calculate the market share of non-state bank deposits. After dominating the Chinese banking sectors for decades, the four leading state-owned banks started witnessing increasing competition from new banks. Firms from areas with greater non state-owned bank activities usually tap more on the independent commercial banks, which closely follow modern banking practices. Now that we have constructed proxies for the three institutional factors, we next construct proxies for different aspects of the distressed firm decisions to separately test our four hypotheses on operating performance, capital structure, investment policy, and recovery outcome. In particular, we intend to take a closer look at these four aspects to gain further understanding about how firms perform (the concurrent operating performance), how managers and investors expect the firms to perform (the forwardlooking capital structure and investment policy), and the resulting outcome of the managerial decisions (the resolution of the cases). To gauge firm performance during financial distress, we focus on three measures, return-on-sales (ROS), total-factor-productivity (TFP), and earnings growth. ROS is defined as a firm s net earnings divided by total sales. To estimate a firm s TFP, we regress total firm sales on two factors, labor and capital, within each industry for each year, to arrive at the Cobb-Douglas production function. We then use the estimated 17

18 industry-level production function to predict a firm s expected production level and take the regression residual as the TFP of the firm. Earnings growth is defined as year-by-year percentage increase in firms net earnings. To track a firm s capital structure, we adopt two measures: leverage, defined as total liabilities debts divided by total assets, and long-term-debt ratio, defined as the ratio of a firm s long term debt to its total debt. We further supplement the capital structure measures with three measures of corporate investment policy: internal investment, external investment, and total investment. Internal investment is defined as a dummy variable that equals to one if there is an increase in a firm s total fixed asset investment from the previous year, and zero otherwise. External investment is defined as a dummy variable that equals to one if external investment in the current year is larger than that of the previous year, and zero otherwise. We define external investment as firm investments in other companies, often through securities investment. Similarly, total investment is a dummy variable if the level of total investment in the current year is greater than that in the previous yearn. Finally, we adopt two variables to track recovery outcome for financially distressed firms. First, we create a binary variable that equals to one if a firm emerges from financial distress and remains out of distress till the end of the sample period, and zero otherwise. In addition, we calculate the number of years that that a firm spends in financial distress, since the beginning and till the end of the period between 2001 and For example, if a firm stays in distress through the sample period, the observation takes the value of 5. In most regression specifications, we include the following independent control variables in addition to the interested variables explained above: 1). financial distress dummy: a dummy variable that equals to one if a firm is classified as in distress (See Appendix II); 2). firm size: the logarithm of company assets in thousands of RMB in the first year of our investigation period; 3). firm leverage: total liabilities divided by total assets in the first year of our investigation period; 4). sales growth: year-by-year total percentage change in sales; 5). tangible asset ratio: the fraction of total tangible assets to 18

19 total assets during the year right before financial distress in the first year of our investigation period; 6). Firm age: the number of years that a firm has been in existence, in the first year of our investigation period. Appendix III provides a summary of the above discussed variables. Table 1 provides summary statistics of these variables. 4. Empirical Results To estimate the impact of different factors on firm behavior in financial distress, we perform panel regression of different aspects of distressed firms behavior on the three different types of institutional factors. To account for any effects of the artificial cut-off that defines our sample period, we perform hazard rate model to estimate the probability of firms emerging from distress and perform tobit model to estimate the time that firms spent in financial distress. Below are the regression results Firm Operating Performance We present results on how ownership structure, government quality, and financial development influence distressed firm behavior. We first examine how such factors influence firm operating performance in Table 2. We regress the three proxies of operating performance (return-on-sales, total-factor-productivity, and earnings growth) on proxies for the three aspects of institutional background, namely ownership structure, government quality, and financial development. The results are highly consistent with the common belief that SOEs suffer from relatively poor performance compared to private firms, reflected by the significantly negative coefficient on the SOE dummy variable. Consistent with prior literature, distressed firms overall suffer poor performance in the ensuing years, compared to the control firms. More interesting to the current paper, it is apparent that SOEs in financial distress perform particularly poorly, in that the coefficient for the interaction between SOE dummy and financial distress is highly significantly negative. Distressed SOE firms suffer about 5 percentages lower ROS than private firms in distress. Some corroborating 19

20 evidence from the continuous variable (state ownership) confirms that higher ownership by the state indeed suggests poorer performance, especially when firms are in distress. One standard deviation increase in state ownership ratio can decrease return-on-sales by about 5 percentages. Such findings provide strong support to our hypothesis that firms with greater state ownership, our proxy for poorer institutional background, perform worse in distress. Our results on the other two proxies for operating performance, total-factorproductivity and earnings growth, generate consistent results. State ownership dummy reduces the TFP by 0.29, about one third of one standard deviation in TFP. In the mean time, state ownership reduces earnings growth by similar magnitude. In addition to the analyses that include one variable of interest at a time in a regression, we perform an alternative specification in which we include three proxies for institutional background, one from each of the three categories (state ownership percentage, government protection, and financial market development) in the same regression. Although some variables become marginally smaller, all three variables come out highly significant in the expected sign, confirming the above results. In sum, there is strong support that institutional background considerably influences distressed firm performance Firm Financing Decisions Next, we test the institutional background s impacts on firms financing decisions in Table 3. Although some of the decisions are driven by liquidity constraints and arguably out of the management control, we believe that management at distressed firms take active control of companies in distress and can indeed take actions, particularly financing decisions, which are critical to the survival and recovery of distressed firms. Therefore, we expect institutional background to have similarly profound impact on firms financing decisions as they have on operation decisions. Our results indicate that average SOEs (both distressed and control firms) indeed take on modestly less debts, relative to their assets and opt for about three percentages of 20

21 more long-term debts to total debts, compared to their private counterparts. This suggests that despite the overall size of the soft lending practice, SOEs are not particularly more indebted than private companies. However, consistent with our hypothesis, we find that distressed firms with greater state ownership take on much greater debt in financial distress than the private companies. An average SOE firm takes on 5 percentages more debt relative to its assets, than an otherwise similar private company. While better government effectiveness (proxied by government quality and the level of market openness) generally provides greater debt capacity to companies, it turns out that distressed firms are less leveraged in those areas. This is again consistent with the argument that better government facilitates contract enforcement and creditor protection. As a result, distressed firms are more motivated to reduce their debts and improve their financial health. The results on financial development are also similar. Better financial development generally enables companies to raise more capital through the relatively cheaper way of debt financing. However, once a company gets in distress, areas with better financial development seems to witness stronger monitoring on the distressed firms, manifested by the distressed firms reduction of debts and the maturity of their debt financing. Our specification that includes all three types of institutional background variables generate qualitatively similar results As to debt maturity, we find that distressed firms with greater state ownership are more likely to increase their long-term debts while firms from areas with better developed financial sectors tend to reduce their long-term debts. Similarly, firms from more financially developed areas take relatively less long-term debts in financial distress, consistent with our hypothesis. The results on government effectiveness on debt maturity are mixed. In an alternative specification in which we include all three categories of institutional background in the same specification, we again find some mixed results: the coefficient on financial development becomes insignificant and the coefficient of government effectiveness indeed turns positive. We are not surprised by the results for at least two reasons. First, extant studies are somehow divided in their predictions/findings on what firms should do with the debt 21

22 maturity in distress (Dinc 2005, Ongena et al. 2003, among others). In addition, we believe that institutional background can indeed yield ambivalent predictions regarding the use of long-term relative to short-term debt. On one hand, better environment impose greater constraints on distressed firms for them to repay their defaulted long-term debts. On the other hand, it is possible that better governmental quality and financial development encourages more debt renegotiation and private workouts, which may lead firms to swap their short-term debts to long-term debts, providing firms the much needed liquidity to completely repay their debts Firm Investment Choices Further, we study how distressed firms adjust investment policies, in response to external institutional environment in table 4. Interestingly, institutions seem to have opposite impact on internal and external investment. SOEs firms and firms from areas with worse governance and financial development witness decreased investment in their existing lines of business during distress. Somewhat surprisingly, this pattern is contrasted by a weak increase in the same firms external investment activities. In addition, we perform regression on the level of total firm investment on institutional variables and the results are mixed and weak. Rather than shifting the total level of investments, our findings seem to suggest that firms facing different institutional background choose different routes to resolve their financial difficulties. Put differently, firms treat internal and external investment like substitutes for each other when it comes to corporate investment decisions. Firms with better governance focus relatively more on company operation and try to improve the performance of existing business, consistent with our prior findings on firm performance in distress. In contrast, firms facing less monitoring and market discipline attempt to navigate through distress by diversifying or switching to new business. This finding is somewhat confirmed by our prior findings that such firms suffer relatively poor performance from operation. This partly reflects the mis-steps that management took before distress and partly confirms our conjecture that such distressed firms put less emphasis on improving existing businesses and instead try to explore new opportunities. Although it is difficult to evaluate the relative merits of the alternative 22

23 strategies, switching from existing to new products apparently involves taking greater risks and possibly takes longer for the firms to repay their debts. Our findings hence support the hypothesis that firms under loose market discipline are more likely to take riskier projects while in distress Recovery from Distress One way to evaluate the relative success of firm strategy within distress is by looking at whether firms manage to emerge from distress. Despite that some SOEs may be protected by government and realize that there is little chance for them to go out of business, our review of the institutional background in Section 2 suggests that even managers at SOEs have incentives of emerging from distress, both for the sake of the firm, and of themselves. We perform the following two analyses in Table 5 to answer this question. First, we perform hazard rate model to examine which types of firms are more likely to emerge from distress at the end of our sample period. For each firm within distress, we trace their performance throughout the rest of the sample period and construct the following dummy variable: if a distressed firm by definition at the start of 2001 is no longer considered distress after 2001 by our criteria and has since stays out of distress during the rest of the sample period, we define the firm as emerging from the distress and the set dummy variable to 1. The dummy variable equals to zero, otherwise. The hazard rate model specification addresses the fact that we have a 6-year observation window for recovery to happen and our observation of the ultimate recovery is truncated. Our results in Table 5 suggest that, other things constant, SOEs are about 20 percentages less likely to emerge from distress by the end of our sample period. On the other hand, firms from areas with better governance quality and financial development are more likely to emerge. Such results, taken together with our previous results on firm decisions, indicate that better institutional background provide greater incentives for firms to take the proper turnaround strategy and that firms are more likely to obtain the objective of improving business and successfully re-structuring. 23

24 In addition, we estimate how institutional factors modify how long it takes firms to emerge from distress. In particular, we calculate the number of years between 2001 and the year that a distressed firm no longer satisfies our distress definition and since stays out of distress. By construct, the minimum number of years is one (a firm emerges from distress in 2002 and has since stayed out of distress) and the maximum number of years is five (a firm stays in distress till the end of our sample period in 2005). We adopt the tobit regression approach to account for the fact that the data may be truncated due to the length of our sample period. Similar to our results on recovery probability, SOEs on average take longer to emerge while firms with better institutional background take less time to emerge, completely consistent with the above findings on recovery probability. Combining the results from the above specification, our evidence seems to suggest that it is more likely and takes less time for firms under stringent governance and discipline to emerge from distress. Our findings seem to support the line of argument (Bai et al. 2002, Brandt and Zhu 2000) that the invisible hand at market place plays a powerful role in shaping firm behavior towards the more efficient direction Further analyses A. Change in distressed firms policies We next perform a host of additional analyses to verify the robustness and gain more understanding about our results. One apparent possibility is that the distressed firms under poor institutional background may have always displayed poor operating performance and aggressive investment policies, even before they enter distress. We perform another two alternative analyses to show that we are not simply picking a firm fixed effect in our main regressions. First, we repeat the same regression as in Tables 2 through 4 with the change in the dependent variables (as opposed to the absolute level of the dependent variables in distress), from pre-distress to post-distress period. In particular, we calculate the average performance for the same firm during the 3-year pre-distress period and the 5-year post-distress period. We then calculate the percentage performance change in each firm and perform cross-sectional regression of performance change on the institutional background. This approach is in essence a difference-in-difference approach, which enables us to pinpoint how firms adjust their decisions after they realize that they 24