The political, economic and social impacts of COVID-19 have attracted significant academic interest (Phan & Narayan, 2020). This paper studies the impact of COVID-19 on firm liabilities in the energy and power industry chain by using a difference-in-difference (DID) model. Our findings confirm that the COVID-19 pandemic is associated with rising debt. We show that firm liabilities for poorly performing firms also significantly increased during the COVID-19 crisis.
Our contributions are as follows: First, contrary to the evolving literature (see Qin et al., 2020; Salisu & Adediran, 2020; Sharif et al., 2020; Shen et al., 2020; Topcu & Gulal, 2020), we turn attention on a different aspect of COVID-19 research: that is, we study the impact of COVID-19 on firm liabilities. None of the studies evaluate the effect on liabilities, which is surprising given that the lack of economic activities has led to business losses and hence have put firms into financial stress. Therefore, it is natural to expect that COVID-19 may have also influenced firm liabilities—to what extent, though, is unknown.
We focus on energy and power industry firms’ liabilities. Our focus on the energy sector connects to the literature on COVID-19 and energy, where researchers have focused on different aspects of the energy industry. Polemis & Soursou (2020) show that stock returns of the Greek energy companies are affected by COVID-19; Salisu & Adediran (2020) show that COVID-19 has contributed to energy market volatility; and Fu & Shen (2020) demonstrate that due to COVID-19 energy companies have performed poorly.
II. Theoretical analysis and hypothesis
Recent years have seen growth of and investments in new energy- related industries. Importantly, innovation in the energy and power industry chain has also grown, and firm liabilities have risen to meet the growth in investments, including potentially excessive investments in oil, nature gas, and human resources (He et al., 2019). Compared with other industries, the debt financing scale of companies in the energy and electric power chain is larger because of its need for infrastructure. However, the sudden COVID-19 outbreak hampered normal operations, with withdrawal of investments, and the burden of high fixed expenses (Chang et al., 2020). Given this situation, firm profitability suffered in the COVID-19 crisis (Shen et al., 2020). Profitable companies overcame funding pressures (Shen et al., 2018).
With energy firms, poor performance implies less access to internal sources of funds. These firms have had to depend on external financing to meet fixed expenditures (Fu & Shen, 2020). Another feature of the debt market has been the role of the stock market volatility which increased due to COVID-19 (Ali et al., 2020) and disrupted innovation-related investment (Shen et al., 2021). Given that the availability and cost of debt financing is lower than equity financing, corporate liabilities have increased during the COVID-19 pandemic. Therefore, we hypothesize that the COVID-19 pandemic caused firm liabilities to increase in the energy and electric power industry.
III. Model and data
In this paper, we take the first quarter financial data of Chinese listed companies in Energy and Power industry chain from 2015 to 2020. We exclude the following samples: (1) distressed firms; (2) finance and insurance firms; and (3) firms with missing data. All data are obtained from the CSMAR database, and continuous variables are winsorized at the 1% and 99% levels. The following DID model was used to test the main hypothesis:
In this model, the asset-liability ratio (LEV) is the dependent variable, which is measured by the ratio of liabilities to total assets and reflects the level of firm liabilities. PERIOD is a dummy variable to signify the period of the COVID-19 outbreak: it takes a value 1 in the first quarter of 2020 and 0 otherwise. SERIOUS is a dummy variable of the “degree of the pandemic impact” based on return on total assets (ROA). When the annual ROA observation is lower than current ROA’s 75th percentile values, it takes a value 1, otherwise we set it to 0. We also control for the following variables: SIZE (calculated by the logarithm of total assets), ROE (the ratio of net profits to net assets), GROWTH (measured by revenue growth to current operating income), FIXS (the ratio of fixed assets to total assets), CASH (the ratio of cash-holdings to total assets), MB (book-to-market equity ratio), and LARGE (share ratio of the largest shareholder). Table 1 describes the descriptive statistical analysis of each variables.
IV. Empirical analysis and results
A. Unit root test and parallel trend test
The unit root test shows that all variables have passed the test at the 1% level, and the integration order of the panel data can be considered stable. The parallel trend test shows that the data from high- and low- impacted companies before the pandemic have the same trend, which can be further studied via the DID modeling. The results are shown in Figure 1.
B. Regression results
Table 1 shows the empirical results. In Column (1), all companies in the energy and power industry are taken as samples to test the impact of COVID-19 on firm liabilities in the energy and power industrial chain. The interaction coefficient of SERIOUS*PERIOD is 0.0278 and significantly positive at the 5% level, indicating that the COVID-19 pandemic had a strong impact on corporate liabilities in the energy and power industry. High fixed cost, matured payments on capital and interests cannot be repaid without strong firm performance and sufficient cash flow. Meanwhile, larger scale and higher proportion of fixed assets can lead to growing debt, and compared to companies with lower profitability, slower revenue growth, and higher cash-holdings, strong financial strength relieves the pressure on corporate debt.
V. Robustness test
In robustness tests, we apply an alternative dependent variable, a propensity score matching approach, and reset samples to evaluate the reliability and stability of the main empirical results. Firstly, the ratio of liabilities to revenue was treated as an alternative dependent variable. The SERIOUS*PERIOUD coefficient is 1.7191 and it is significantly positive at the 1% level. The results remained consistent (see Column 2, Table 2). Secondly, the companies in the energy and power industry are taken as the experimental group to compare with other industry companies in the DID test, and propensity score matching was used to overcome sample selection bias. The empirical result in column (3) remains robust and the model is as follows with all variables are defined previously in Equation (1):
This paper studies firm liabilities in the energy and power industry chain. We find that the liabilities of companies in the energy and power industry chain increased significantly during the COVID-19 pandemic period. On the one hand, the external market environment was more uncertain in the COVID-19 crisis, thus production and consumption in the energy sector were seriously impacted. On the other hand, high fixed expenses, poor performance and tight cash flows led to growing debt and disinvestment in the energy industry.
We acknowledge support from Science and Technology Department of Guizhou Province (4Y173) in the completion of this project. We would also like to thank the anonymous referee and Editor of Energy RESEARCH LETTERs for their valuable comments in improving this study.