Economic and environmental sustainability are crucial for the long-term sustainable development of corporations (Wu et al., 2020). While accounting conservatism is widely recognized in improving the transparency of companies' financial information (Ruch & Taylor, 2015) and enabling effective contracting with various stakeholders (Guo et al., 2020). The Chinese government intervenes in financial markets and plays a key role in the economy (Cull et al., 2015).
Although existing studies have examined the relationship between accounting conservatism and corporate social responsibility (Anagnostopoulou et al., 2021; Cho et al., 2020), there are still unresolved controversies. Our study extends the literature that accounting conservatism not only improves the quality of corporate financial reporting (Ruch & Taylor, 2015), but also facilitates communication with various stakeholders (Guo et al., 2020). Unlike Anagnostopoulou et al. 2021) that examine conflicts between managerial self-interest and stakeholders.
According to Zhang et al. 2016), political connections between companies and governments can be divided into two types. More importantly, they have better personal channels to communicate with government officials (Hillman et al., 1999).
Data and methodology
Resource dependence on the government makes firms more likely to comply with the government's requirements and even put extra effort into socially identifiable behaviors to gain government favor (Pfeffer & Salancik, 1978). We expect that firms subject to heavy regulatory constraints are more likely to commit to environmental responsibilities. Therefore, the negative correlation between accounting conservatism and environmental performance will be attenuated in those firms.
Referring to Sarfraz et al. 2020) we produce a comprehensive measurement of environmental disclosure, labeled as Disclosure, which covers four main dimensions, namely legal awareness, social valuation, environmentally friendly production and green management.3 Disclosure data is collected from the CSMAR database. We then sum the scores of the 11 items as the environmental disclosure variable labeled Disclosure. Referring to the method of Li et al. 2021), we adjust this variable for the industry year.
Finally, according to Akbar et al. 2021), as a fourth measure of environmental performance, we use environmental investment, denoted by EI, which is calculated as the natural logarithm of environmental investment plus one, with data collected from the CSMAR database. Unlike environmental expenditures, a firm's investment behavior is a function of expected future returns (Khan et al., 2016). The measure of accounting conservatism adopted in this paper is the level of annual accounting conservatism developed by Khan & Watts (2009).
Then, by plugging them into equation (3), the fixed-year conditional accounting conservatism measure CScore (𝛽3) can be calculated, which is the measurement of accounting conservatism in this study. We control for other variables that may affect environmental performance based on the literature (Anagnostopoulou et al., 2021; Cho et al., 2020; Pan & Zhao, 2021).
Using equation (4), coefficients 𝜔0, 𝜔1, 𝜔2 and 𝜔3 can be estimated by regression with the annual cross-sectional data. To address the unobserved firm characteristics and time-varying heterogeneity, we control for fixed and year fixed effects.
We adopt a two-stage least square (2SLS) approach to explore the role of short-term creditors. preference in the relationship between accounting conservatism and environmental performance. Therefore, we suspect that high accounting conservatism will reduce analyst attention, and therefore, firms choose to reduce environmental engagement accordingly. Therefore, we hypothesize that analyst attention is a channel to explain the substitution effect of accounting conservatism on environmental performance.
Similarly, we use a 2SLS approach to examine whether analyst attention explains why accounting conservatism can substitute for environmental commitment. 2012) and Piotroski and Roulstone (2004), we construct two variables to measure the degree of analyst attention. The results in Panels A and B of Table 4 (first-stage analyses) show that accounting conservatism is negatively related to analyst attention or report attention. We run regressions adding the interaction term of CScore*Described bureaucratic connection and respectively CScore*Achieved political connection, and the results are shown in Table 5. The results show that attributed bureaucratic connection can increase the substitution effects of accounting conservatism and environmental performance, demonstrating H2a.
In contrast, the negative relationship between accounting conservatism and environmental performance becomes insignificant when firms with executives have achieved political connections, which is consistent with H2b. Government monitoring of pollution control and environmental protection will directly affect the certification of the operation of highly polluting firms (Matsumoto & Szidarovszky, 2020; Shahbaz et al., 2015), thus changing the attitude of firms towards greater environmental participation . that under policy adjustment pressure, the negative relationship between accounting conservatism and environmental performance will become insignificant. The result is consistent with H2c that when faced with policy constraints, managers should continuously engage in environmental activities to reduce the risk of litigation, and therefore, the substitution effect of accounting conservatism disappears.
In this section, we investigate the heterogeneous factors that influence the impact of accounting conservatism on environmental performance. Therefore, we suspect that the proxy effect of accounting conservatism may vary depending on the magnitude of the firm's financial constraints, thereby influencing the relationship between accounting conservatism and environmental performance. Panel A, Panel B, and Panel C show that the relationship between accounting conservatism and environmental performance becomes insignificant in firms with high financial constraints.
Therefore, we expect that in a highly competitive industry, managers will not reduce environmental performance even if they have adopted accounting conservatism. A and B in Table 8 show that when industry competition is high, the negative correlation between accounting conservatism and environmental performance becomes insignificant. The results confirm our expectation that managers intend to engage more in green activities when they face intensive competition, as such the substitution effect of accounting conservatism becomes insignificant.
We note that the coefficients of Fitted_CScore are negative and statistically significant at the 5% level in both columns. The Hansen's J statistics are not significant and indicate that the null hypothesis that the instruments are valid cannot be rejected. The baseline regression takes into account the influence of unobserved firm characteristics and time-varying heterogeneity, thus controlling for firm and year fixed effects.
We add industry×year fixed effects and province×year fixed effects in Table 10 to address the concern that unobserved firm characteristics and time-varying heterogeneity across industries and provinces may affect corporate environmental involvement. The results in Table 10 show that accounting conservatism is still significantly and negatively associated with environmental performance indicators after controlling for multiple fixed effects. We group firms that receive at least one fine due to financial fraud as treatment firms, while those that do not receive a financial fraud-related fine in a year as the control group.
That is, receiving a fine due to financial fraud in year t-1 is approved as a hit, which increases accounting. According to Wei (2021), firms will increase their accounting conservatism to maintain legitimacy after financial reporting anomalies or relative penalties. By taking the shock as an exogenous event of improving accounting conservatism, DID estimation helps justify causality.
It shows that the coefficients of Post*Treat are negatively and significantly related to the two measures of environmental performance, indicating that firms reduce their environmental commitment after a punishment shock. This result supports our baseline finding that an increase in accounting conservatism will lead to a decrease in environmental commitment. To avoid measurement error, we use two alternative environmental performance indicators to reconfirm our baseline results.
We use cost/sales, which is the expenditure on environmental involvement divided by sales and multiplied by 1000, as the third environmental involvement measurement. 2021), this variable was adjusted operating year. Corporate environmental investment, denoted as EI, is calculated as the natural logarithm of corporate environmental investment plus one. 2021), we use environmental investment (EI) as the fourth variable to measure environmental performance. The results in Table 12 show that accounting conservatism is significantly and negatively associated with the adopted alternative measures of environmental performance.
In addition, internal and external pressures, such as financial constraints and industry competition, also moderate the substitution effect of accounting conservatism. Our results highlight that the adoption of accounting conservatism can function as an efficient mechanism influencing corporate environmental commitment. An empirical investigation of the relationship between change in corporate social performance and financial performance: a stakeholder theory perspective.
Big4 A dummy variable equals one if the firm's accountant is one of the international 'Big4' accounting firms, and zero otherwise. GDP Growth Rate The GDP growth rate per capita of the province where the company is located. HHI The Herfindahl-Hirschman index that reflects industrial concentration, calculated as the sum of the squares of the market shares of all companies in a given market.
Analyst attention The natural logarithm of the number of analysts (teams) covering a company in a year. Report Attention The natural logarithm of the number of analytical research reports on a company in a year. It reports the impact of accounting conservatism on two measures of a firm's environmental performance, including Hexun's industry-year-adjusted environmental responsibility (ER) score and environmental disclosure score (Disclosure).
This table presents the 2SLS regression results using analyst attention and report attention as the mechanism through which accounting conservatism affects firm environmental performance. Analyst attention is the natural logarithm of the number of analysts (teams) covering a company in a year. Report attention is the natural logarithm of the number of analyst research reports on a company in a year.
This table reports the regression results on whether political connection moderates the relationship between accounting conservatism and corporate environmental performance. Columns (1) and (2) present the results of the regression controlling for firm and operating year fixed effects. Columns (3) and (4) present the results of the regression controlling for fixed and province-year fixed effects.
Receipt of financial fraud fine in year t-1 is adopted as a shock that increases accounting conservatism but affects environmental performance directly, to justify the causal relationship between accounting conservatism and environmental performance.