This study aimed to evaluate whether the stimulus package after the 2008 financial crisis led to the distortion of financial intermediation in China. Accordingly, we investigated the relationship between company survival and the financial intermediation function in China during the sample period, which reflects the financial crisis in 2008. The study presents two main findings. First, access to bank loans resulted in a decrease in the risk of failure for the relatively poorly performing state-owned enterprises (SOEs) in China after the 2008 financial crisis. Distorted financial intermediation by banks was more pronounced for the large-sized SOEs. Therefore, “zombie firms” inevitably emerged among the manufacturing SOEs in China as result of the financial crisis and subsequent fiscal expansion. Second, access to short-term bank loans led to a deterioration rather than an improvement in the performance of the poorly performing SOEs after 2007.
This study has not adopted the approach that uses an a priori definition of zombie firms, such as that utilized by Hoshi (2006), Caballero, Hoshi, and Kashyap (2008), Shen and Chen (2017), McGowan et al. (2017), Lam et al. (2017), and Faccio et al. (2005), as such an approach almost automatically detects zombies in the sample and, thus, cannot provide an adequate answer to the research question as to whether the financial crisis in 2007 truly gave birth to zombie firms. However, to empirically investigate the emergence of zombie firms in China, we also checked whether bank loans work as a form of relief finance for the firms with negative equity value. It should be noted that the “firms with negative equity value” as a factor comprising a definition of zombie firms stems from the definition of zombie firms utilized by these authors.
To be more detailed, the third quartile firms in performance measures such as profitability (ROA) are “relatively poorly performing” ones, which can be seen as the firms having the possibility of surviving with access to bank loans. Firm size was measured based on the total assets held by firms.
Although this work essentially presents a form of analysis of firms’ failure and survival, a formal survival analysis, such as Cox’s proportional hazard model, has not been employed.
This study differs from Tsoukas (2011) and other studies investigating firms’ survival within this strand, in that we did not necessarily hold a negative view of increases in the likelihood of failure for firms. Rather, we think it reasonable for firms to have a higher failure hazard if they have larger amounts of bank borrowing as opposed to obtaining their funds through issuing stock. A larger amount of debt increases the risk of default and bankruptcy, whereas issuing stock only leads to an increase in the firm’s own capital.
If we include long-term bank loans as another leverage variable into the Cox proportional hazard model, the changing specifications occasionally lead to unstable estimation results. Also, in the estimation of the profit function explained later, if we use long-term bank loans as an independent variable, stable estimation results cannot be obtained. Therefore, we did not use long-term bank loans as another leverage variable. These unstable estimation results are likely to be because the long-term bank loans variable takes the value of zero in most of the observations in the data used by this study.
We test the hypothesis that the coefficient is greater than one, or less than one, and the test statistic would be greater than zero, or less than zero.
In the construction of the conceptual framework, our profit function model analysis used a sample that comprised only those firms that survived in market. This analysis can be insightful as it is complementary to the survival analysis using the hazard model.
The estimation results for μj, and μp are not reported due to space limitations μi could not be estimated due to the nature of the system generalized method of moments estimation technique that we adopted.
We used two-step GMM instead of one-step GMM since the former is asymptotically more efficient. Therefore, we applied the Windmeijer (2005) finite sample correction to the two-step covariance matrix to settle the potentially downward-biased two-step standard errors.
In this study, we defined the disappearance from the Chinese NBS dataset as the exit of a certain firm. This definition is reasonable, because the datasets were obtained from a census, although a very small fraction of firms disappearing from the Chinese NBS dataset actually decreased their annual sales from more to less than five million yuan. This check was conducted using the Qin database that was provided by Bureau van Dijk. That database covers the firms with even smaller annual sales than five million yuan.
We confirmed that the firms drop** out of our data sample did not disappear because they merged with other firms or changed their names from one year to another; these firms only disappeared from the markets.
Furthermore, to check whether the SOEs behave in a different way from one area to another in China, we tested it by including the interaction terms of the important independent variables and coastal area dummy into our empirical models (the inland area is the reference). It was found that the SOEs in the same subgroup in terms of their profitability and/or size behave in largely the same way as those in the inland provinces, although for brevity, the results are not reported in the tables.
To ensure that our findings are also relevant for other countries, we tried to conduct additional empirical practices in this study using the absolute levels of the firm size and firm profitability standards to define large-sized firms and poorly performing firms. This is because, according to the definition of large-sized firms and poorly performing firms based on the quartiles classification, the absolute levels of large-sized firms and poorly performing firms vary from one country to another. Specifically, we adopted the firms with more than 5,000 employees and with negative ROA during our sample period on average as the absolute level definitions of large-sized firms and poorly performing firms, respectively. In addition, we used 3,000 employees and ROA of less than minus five percent as alternative thresholds dividing firms into the large-sized and non-large sized ones and well performing and poorly performing ones, respectively. The results from the Cox proportional hazard model are reported in Table 9, and these were similar to the baseline results shown in Table 4. Also, in any other trial using these absolute levels of firm size and firm profitability standards, we obtained qualitatively similar results to those shown in this study, although for brevity, we do not report them here.
The gap for the coefficients between the periods before and after 2007 was statistically insignificant for large-sized firms in the fourth quartile in terms of profitability. We suppose that the banks had some tendency to hesitate to bail out the poorest-performing SOEs in the fourth quartile in terms of profitability after 2007, which were risky borrowers for the banks, even though these SOEs are large-sized ones. On the one hand, the pressure from the government for the banks to bail out the poorly performing SOEs made the banks embark on the provision of inefficient loans for the moderately poorly performing SOEs in the stimulus package scheme after 2007. On the other hand, even the pressure from the government could not persuade the banks to do the same for the poorest-performing SOEs.
On the contrary, for well performing SOEs, the first and second quartiles group in terms of the profitability, the effects of access to short-term bank loans on their performance significantly converted from a deterioration to an improvement before and after 2007.
Another firm performance measure is Added value/ Total assets; that is, productivity or, more specifically, capital productivity. Even when using Δ(Added value/ Total assets) instead of ΔROA, we checked whether the bank loans led to an improvement in firm performance; similar observations were obtained, although they are not reported here to save space.
There is a caveat to the interpretation of the results. In Table 6, we should note that the conversion from an improvement to a deterioration in their performance before and after 2007 was not observed for the fourth quartile in terms of profitability in the largest-sized SOEs. For them, a decrease in the risk of failure as a result of short-term bank loan access could be observed clearly after 2007 in the survival analysis using the Cox proportional hazard model in Table 4. For these SOEs, the coefficients of Short-term loans /Total assets are positive for the periods both before and after 2007, although they are insignificant. We have to admit that the stimulus packages in terms of bank loans did not only simply bail out these SOEs but also improved their performance after 2007 to some extent. The improving effect due to the new bank loans enhancing the firms’ profitability and the deteriorating effect due to the bailout loans spoiling their profitability were likely to result in the insignificant improvement in the performance of these SOEs after 2007. An anonymous referee pointed this out. We are grateful for this valuable comment.
References
Audretsch, D., and T. Mahmood. 1995. New firm survival: New results using a hazard function. The Review of Economics and Statistics 77: 97–103.
Bridges, S., and A. Guariglia. 2008. Financial constraints, global engagement, and firm survival in the UK: evidence from micro data. Scottish Journal of Political Economy 55: 444–464.
Bunn, P., and V. Redwood. 2003. Company accounts based modelling of business failures and the implications for financial stability. Bank of England Working Paper 210.
Caballero, R.J., T. Hoshi, and A. Kashyap. 2008. Zombie Lending and Depressed Restructuring in Japan. American Economic Review 98: 1943–1977.
Chen, M., and A. Guariglia. 2013. Internal financial constraints and firm productivity in China: Do liquidity and export behavior make a difference? Journal of Comparative Economics 41: 1123–1140.
Clark, G., R. Cull, and G. Kisunko. 2012. External finance and firm survival in the aftermath of the crisis: Evidence from Eastern Europe and Central Asia. Journal of Comparative Economics 40: 372–392.
Giannetti, M., and S. Ongena. 2009. Financial Integration and Firm Performance: Evidence from Foreign Banks Entry in Emerging Markets. Review of Finance 13: 181–223.
Görg, H., and M. Spaliara. 2009. Financial health, exports and firm survival: a comparison of British and French firms. Centre for Economic Policy Research (CEPR) Working Paper: 7532.
Guariglia, A. 2008. Internal financial constraints, external financial constraints, and investment choice: evidence from a panel of UK firms. Journal of Banking and Finance 32: 1795–1809.
Haas, R., and N. Horen. 2011. Running for the exit: international banks and crisis transmission. European Banks for Reconstruction and development Working Paper: 124.
Hoshi, T. 2006. Economics of the living dead. The Japan Economic Review 57: 30–49.
Iwasaki, I. 2014. Global financial crisis, corporate governance, and firm survival: The Russian experience. Journal of Comparative Economics 42: 178–211.
Lam, R., A. Schipke, Y. Tan, and Z. Tan. 2017. Resolving China’s zombies: Tackling debt and raising productivity. IMF Working Paper: 17/266.
Liu, Q., X. Pan, and G. Tian. 2018. To what extent did the economic stimulus package influence bank lending and corporate investment decisions? Evidence from China. Journal of Banking and Finance 86: 177–193.
McGowan, Adalet, M., D. Andrews, and V. Millot. 2017. The walking dead? Zombie firms and productivity performance in OECD countries. Economics Department Working Papers: 1372.
Roodman, D. 2008. How to do xtabond2: An introduction to difference and system GMM in Stata. Center for Global Development Working Paper: 103.
Shen, G., and B. Chen. 2017. Zombie Firms and over-Capacity in Chinese Manufacturing. China Economic Review 44: 327–342.
Shiferaw, A. 2009. Survival of Private Sector Manufacturing Establishments in Africa: The Role of Productivity and Ownership. World Development 37: 572–584.
Spaliara, M. 2009. Do financial factors affect the capital-labour ratio? Evidence from UK firm-level data. Journal of Banking and Finance 33: 1932–1947.
Spaliara, M., and S. Tsoukas. 2013. What matters for corporate failures in Asia? Exploring the role of firm-specific characteristics during the Asian crisis. Structural Change and Economic Dynamics 26: 83–96.
Tsai, K. 2015. The political economy of state capitalism and shadow banking in China. HKUST Institute for Emerging Market Studies (IEMS) Working Paper: 2015-25.
Tsoukas, S. 2011. Firm survival and financial development: evidence from a panel of emerging Asian economies. Journal of Banking & Finance 35: 1736–1752.
Wu, W., C. Wu, C. Zhou, and J. Wu. 2012. Political connections, tax benefits and firm performance: Evidence from China. Journal of Accounting and Public Policy 31: 277–300.
Yang, J., A. Guariglia, and J. Guo. 2019. To what extent does liquidity affect M&As’ method of payment and performance? Evidence from China. Journal of Corporate Finance 54: 128–152.
This work was funded by Scientific Research (C) No. 19K01633 and Scientific Research (C) No. 20K01630 from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MECSST), The Murata Science Foundation, and The Mitsubishi Foundation No. 30230. The authors acknowledge comments and suggestions from Ichiro Iwasaki, Gang Xu, Yingying Fang, and two anonymous referees. All views and errors remain our own.
Author information
Authors and Affiliations
Faculty of Foreign Studies, The University of Kitakyushu, 4-2-1, Kitagata, Kokuraminami-ku, Kitakyushu, 802-8577, Japan
Moho Shiraishi
Graduate School of Economics, Kyoto University, Yoshida-honmachi, Sakyo-ku, Kyoto, 606-8501, Japan
Shiraishi, M., Yano, G. The Financial Crisis in 2008, the Stimulus Package, and Distortion of Financial Intermediation in China: A Survival Analysis Approach.
Comp Econ Stud64, 280–323 (2022). https://doi.org/10.1057/s41294-021-00165-0
