The Impact of Environmental Information Disclosure on the Efficiency of Enterprise Capital Allocation

Abstract

Environmental information disclosure has become a widely-used tool to encourage the participation of multiple market players in environmental governance. However, it remains unclear whether it can promote the efficiency of capital allocation in enterprises. This study uses econometric modeling and data from heavily polluting enterprises in Chinese A-shares between 2013 and 2020 to explore the impact of environmental information disclosure on capital allocation efficiency, as well as its mechanisms. It is found that environmental information disclosure significantly and robustly enhances the efficiency of capital allocation, and the effect varies by firm’s size, ownership, life cycle, and region. Nevertheless, employees and creditors are found to have a negative moderating role in this effect. These findings have important implications for the simultaneous improvement of environmental performance and capital allocation efficiency in the context of China’s ecological civilization system and high-quality economic development and for promoting a “win-win” situation for environmental protection and economic growth.

1. Introduction

Enterprises are the main production entities in the market, and improving their capital allocation efficiency is crucial for achieving profit maximization and sustainable development. With increasing global attention on environmental governance, many companies use environmental information disclosure to demonstrate their sustainable development capabilities [1,2]. Environmental information disclosure is an important non-financial information that provides a valuable reference for enterprise managers and external investors, influencing their decision-making processes [3].

Despite a considerable body of research focusing on the relationship between environmental information disclosure and enterprises’ economic performance, scholars have not reached a consensus. Some scholars posit a positive promotion effect between the two, where improved quality and content of environmental information disclosure leads to the good economic performance of enterprises. For instance, Dhaliwal et al. (2011) found that proactive corporate social responsibility disclosure reduces equity capital costs, forecasting errors, and profit forecast dispersion [4]. Similarly, Plumlee et al. (2015) reported that voluntary environmental information disclosure increases future cash flows, reduces equity capital costs, and improves economic performance [5] Additionally, Fan et al. (2020) and Liu et al. (2021) also found that environmental information disclosure has a positive impact on corporate profits and financial performance, respectively [6,7].

However, some scholars argue that excessive environmental information disclosure has a negative impact on enterprises’ economic performance due to high disclosure costs. Enterprises spend significant time and resources on the disclosure process, with longer return cycles. Investors prioritize profitability over environmental governance, and excessive disclosure may reduce investor confidence and increase operating risks, ultimately leading to a decline in economic performance. For instance, Chen et al. (2018) analyzed the relationship between environmental information disclosure and enterprise profitability using a difference-in-differences model and found that an increase in environmental information disclosure costs reduces profitability due to increased environmental protection expenditure [8]. Similarly, Ren et al. (2020) found that increased disclosure costs led to a reduction in operating capital, and ultimately, economic performance [9].

Despite the extensive literature on the impact of environmental information disclosure on economic performance, the role of capital allocation efficiency in this relationship has received limited attention. With China’s economic growth shifting from extensive to intensive [10], exploring the role of environmental information disclosure in improving capital allocation efficiency in heavily polluting industry enterprises is crucial for promoting green development. To address this gap, this study uses ordinary least squares regression analysis on panel data from heavily polluting enterprises listed on A-shares in China from 2013 to 2020. The findings of our study demonstrate that enhancing the quality of environmental information disclosure significantly promotes the efficiency of capital allocation in enterprises. Furthermore, we discovered that stakeholder pressure negatively and significantly weakened the relationship between environmental information disclosure and capital allocation efficiency. Additionally, the impact of environmental information disclosure on capital allocation efficiency varies depending on the enterprise’s size, property rights, region, and growth cycle.

Compared to previous studies, our paper’s contributions are threefold. Firstly, we focus on the economic consequences of environmental information disclosure on capital allocation efficiency in enterprises, which is an area that has received relatively little attention in previous studies. Secondly, our empirical analysis reveals that some stakeholders have a moderating effect on the relationship between environmental information disclosure and capital allocation efficiency in enterprises. These findings enrich and expand the existing theoretical research on this topic. Thirdly, we broaden the research field by examining how the impact of environmental information disclosure on capital allocation efficiency varies across enterprises with different characteristics, such as size, ownership, life cycle, and region.

2. Literature Review and Research Hypothesis

2.1. The Relationship between Environmental Information Disclosure and the Efficiency of Enterprise Capital Allocation

In recent years, there has been increasing scholarly attention to the influence of environmental information disclosure on firms’ economic performance. The prevailing research findings suggest that environmental information disclosure has a positive effect on various indicators of firms’ financial performance, such as Tobin’s Q, return on assets (ROA), and stock prices [11,12,13,14,15]. This positive relationship can be attributed to several factors. Firstly, although addressing environmental pollution issues requires significant resource allocation and may lead to short-run cost increments, firms that engage in effective environmental protection measures are more likely to reap long-run financial benefits and higher profits [16]. Additionally, transparent and comprehensive environmental information disclosure enhances investors’ risk tolerance and boosts consumer confidence in firms, which, in turn, mitigates negative evaluations from stakeholders and reduces the overall corporate risk [17,18,19]. Despite the growing body of research in this area, there is a noticeable gap in understanding the impact of environmental information disclosure on firms’ capital allocation efficiency. Capital allocation efficiency serves as a critical indicator of firms’ financial performance, yet it has received limited scholarly attention in relation to environmental information disclosure. This study aims to bridge this gap by conducting a systematic investigation into the relationship between environmental information disclosure and firms’ capital allocation efficiency.

Legitimacy theory suggests that when a company’s legitimacy status is compromised due to environmental issues, environmental information disclosure can be used to influence stakeholders’ perception of the company’s legitimacy [20]. Environmental information disclosure can also reduce information asymmetry both inside and outside the company, helping external investors to gain timely insights into the company’s operational status and reducing inefficient investment behavior [21]. Numerous studies have shown that information disclosure can promote capital allocation efficiency. Good quality environmental information disclosure can help companies establish a positive environmental image, enhance public recognition, and strengthen their reputation capital [22]. Moreover, from the perspective of the information asymmetry theory, environmental information disclosure can help reduce investors’ investment risks and form risk premium compensation, which can help lower a company’s equity capital cost and financing cost, thereby enhancing its capital allocation efficiency [23].

H1. 

Environmental information disclosure has a positive effect on the efficiency of corporate capital allocation.

2.2. The Differential Impact of Environmental Information Disclosure on the Efficiency of Enterprise Capital Allocation

In reality, the decision-making process of companies regarding environmental protection is influenced by various characteristics of the company. This leads to different priorities in the company’s environmental behavior. Therefore, in the process of the impact of environmental information disclosure on corporate capital allocation efficiency, the different attributes of the company can cause differences in this influence. For instance, state-owned enterprises bear more policy burdens than non-state-owned enterprises in China, making them more proactive in adopting environmental protection behavior and disclosing environmental information. This may result in differences in the impact of environmental information disclosure on corporate capital allocation efficiency due to differences in the ownership of the company. This article analyzes the differentiated impact of environmental information disclosure on corporate capital allocation efficiency from four aspects: enterprise size, ownership, lifecycle, and the region where the enterprise is located.

2.2.1. Enterprise Size

The size of a company determines its ability to allocate resources efficiently. Large companies have established management systems and high costs associated with making adjustments, as well as specialized assets that limit their efficiency in resource allocation compared to small and medium-sized enterprises (SMEs). In other words, large companies face higher costs when making information disclosure decisions compared to SMEs. Conversely, SMEs are more efficient in resource allocation, allowing them to quickly allocate multiple resources to reduce the proprietary cost of environmental information disclosure [24]. According to the voluntary disclosure theory, companies should only disclose information that generates benefits greater than their proprietary cost [25,26]. Within this framework, the quality of environmental information disclosure by SMEs is higher than that of large companies. If environmental information disclosure has a significant impact on the capital allocation efficiency of a company, this impact will vary based on the size of the enterprise.

2.2.2. Enterprise Ownership

According to the legitimacy theory, the production and operation activities of enterprises should be recognized by the public and be within the norms of society. Early research suggested that the responsibility of the enterprise was to maximize the interests of shareholders, making company profits an important standard for measuring the legitimacy of the enterprise [27]. However, with the development of the research, some scholars have proposed that enterprises not only need to meet the expectations of shareholders but also need to consider the rights of the public, so they must satisfy the diversified needs of society to maintain their legitimacy status [28,29,30]. In China, state-owned enterprises have a special status and advantages, enjoying more policy and financial support than non-state-owned enterprises [31]. Therefore, the information disclosure decisions of state-owned enterprises do not depend on environmental performance. It leads the initiative of state-owned enterprises in environmental information disclosure will be lower than that of non-state-owned enterprises, which will further lead to differences in the impact of environmental information disclosure on the efficiency of capital allocation due to the nature of corporate property rights.

2.2.3. Enterprise Life Cycle

According to life cycle theory, the life cycle of an enterprise basically follows the three stages of “growth—maturity—recession”, and there are significant differences in the capabilities of enterprises in different life cycles, which may lead to the differential impact of enterprises environmental information disclosure on the efficiency of capital allocation. Specifically, growing enterprises have clear development goals and urgent capital needs, in order to maintain a good image, they will provide higher-quality environmental information to alleviate external financing constraints. In addition, they are sensitive to policy changes and can quickly deploy environmental policies. By comparison, despite mature enterprises having strong profitability and financing capacity, they are not as active in disclosing environmental information as growing enterprises because there is a strong asset specificity among mature enterprises and resulting in an inability to respond flexibly to various emergent environmental policies [32]. For enterprises in recession, there are many difficulties such as declining in sales, market share, and profits, as well as institutional rigidity that negatively affects its production and business activities. Therefore, the main task of those enterprises is to address survival issues rather than respond to environmental risks. Differences in an enterprise’s life cycle also impact its environmental information disclosure, affecting its capital allocation efficiency.

2.2.4. Region Where Enterprise Is Located

According to the environmental Kuznets curve theory, the relationship between economic growth and environmental pollution is an “inverted U”, which means that when the economic growth of a region reaches a particular stage, local governments will gradually pay attention to environmental management issues and thus improve the environmental quality of the region. In China, as the economic development of the eastern and central regions is faster than the western and northeastern regions, the former two regions focus on economic growth and environmental issues, while the western and northeastern regions still focus on economic growth. As a result, enterprises in the eastern and central regions are more environmentally friendly than in the western and northeastern regions. The quality of environmental information disclosure varies due to the different production and business models of enterprises in different regions, which may lead to varying effects of environmental information disclosure on the efficiency of capital allocation of enterprises in different regions.

According to the above analysis, the following hypotheses were formulated for this study.

H2. 

Environmental information disclosure can promote capital allocation efficiency in SMEs.

H3. 

Environmental information disclosure can promote capital allocation efficiency in non-state-owned enterprises.

H4. 

Environmental information disclosure can promote capital allocation efficiency in growth stage enterprises.

H5. 

Environmental disclosure can promote efficient capital allocation efficiency for enterprises in the eastern and central regions.

2.3. Mechanisms of the Impact of Environmental Information Disclosure on the Efficiency of Enterprise Capital Allocation

According to stakeholder theory, there is an interaction between organizations and its stakeholders [33,34]. Thus, there is a potential moderating role of stakeholders in the process of environmental disclosure affecting the efficiency of corporate capital allocation. Stakeholders invest their capital into enterprises for capital appreciation. Therefore, stakeholders must rely on and interact with each other. The joint action of many stakeholders enhances the efficiency of capital allocation in an enterprise. In the process of environmental information disclosure affecting the efficiency of enterprise capital allocation, it is necessary to obtain stakeholders’ approval and thus maintain a good relationship between the various stakeholders. Creditors and shareholders, as the primary source of monetary capital for an enterprise, need to add value to their capital while maintaining the value of their original monetary capital. The interest of such stakeholders in environmental information disclosure is primarily for profit-making purposes. By focusing on the non-financial performance information of the enterprise, shareholders and creditors can further understand whether the operation of the enterprise is conducive to the appreciation of their monetary capital and thus make decisions on whether to continue to provide financial support. Employees, as the primary source of investment in the company’s human capital, are primarily interested in non-financial information about the company for job promotion and salary advancement. When the quality of an enterprise’s environmental disclosure is high, the enterprise also tends to invest more resources in environmental activities, which can crowd out resources that the enterprise uses to enhance the treatment of employees and the capital value of shareholders and creditors, leading to boycotting behavior by these stakeholders.

Based on the above analysis, the following hypotheses are formulated in this paper.

H6. 

There is a negative moderating effect of stakeholders in the role of environmental information disclosure on the efficiency of enterprise capital allocation.

3. Methodology

3.1. Data Sources and Processing

This paper uses panel data of listed companies of heavy polluting in A-shares from 2013–2020 as a sample, the reasons are as follows. (1) Industrial pollution has now become a bottleneck limiting the sustainable development of China’s economy. Enterprises in heavily polluting industries are essential pillars of China’s economic development. They account for 40–50% of total industrial pollution [35], with apparent characteristics such as excessive pollutant emissions and massive resource consumption. (2) Compared with enterprises in other industries, enterprises in the heavily polluting industries disclose more environmental information to the public, so the research sample focuses on enterprises in the heavily polluting industries, and the research is more relevant. (3) Since 2013, China’s socialist construction has entered a new era, which has placed new demands on the business development model, and limiting the time interval of the study sample to after this point in time eliminates the influence of other factors.

According to the classification of heavily polluting industries in the Management List of Environmental Protection Verification Industries for Listed Companies issued by the Ministry of Environmental Protection of China in 2008 and the Guideline on Industry Classification of Listed Companies issued by the China Securities Regulatory Commission in 2012, enterprises in 14 industries such as extractive industries and food and beverage industries were classified as heavy polluting enterprises. In terms of data processing, to eliminate potential bias in the estimation results due to abnormalities in enterprise operations and other circumstances, all enterprises containing ST and *ST, as well as data with outliers or missing values, were further eliminated from the sample in this paper. The sample data of listed companies in China’s A-share heavy pollution industry from 2013 to 2020 was finally obtained, which contained unbalanced annual panel data for 590 firms from 2013–2020 surface, with a total of 4230 observations. The environmental information disclosure data in this paper is obtained from the companies’ annual reports and CSR reports. In contrast, other data is obtained from the CSMAR and the Wind databases.

3.2. Variable Definitions

3.2.1. Explained Variable

The explained variable in this paper is the efficiency of the enterprise’s capital allocation, which is meant to be the whole chain of capital processes within the enterprise, from raising capital upfront to distributing profits later. Considering that the Richardson model can measure the annual capital allocation efficiency of enterprises more intuitively, the expected investment model of Richardson (2006) is chosen to estimate the capital allocation efficiency of enterprises in this paper, and the Richardson model is shown in Equation (1).

$$\begin{array}{c}In{v}_{i,t}={\alpha }_0+{\alpha }_{1}Tobin{q}_{i,t-1}+{\alpha }_{2}Le{v}_{i,t-1}+{\alpha }_{3}Cas{h}_{i,t-1}+{\alpha }_{4}Ag{e}_{i,t-1}+\\ {\alpha }_{5}Siz{e}_{i,t-1}+{\alpha }_{6}Re{t}_{i,t-1}+{\alpha }_{7}In{v}_{i,t-1}+{\epsilon }_{i,t}.\end{array}$$

(1)

In Equation (1), the explained variable Inv denotes the amount of new investment by enterprises. Considering that the existence of underinvestment or overinvestment in an enterprise is not expressed in the current period and acts more on the future economic performance of the enterprise [36]. Therefore, the explanatory variables in the equation are taken as lagged one-period values. $Tobinq$ represents the enterprise’s Tobin’s Q, expressed as the ratio of the market value of equity to the enterprise’s net assets; $Lev$ represents the enterprise’s gearing ratio, expressed as the ratio of total liabilities to total assets. $Cash$ represents the firm’s monetary cash holdings, expressed as the ratio of monetary funds to total assets; $Age$ represents the firm’s years on the market, expressed as the difference between the sample year and the year of listing; $Size$ represents the firm’s asset size, expressed as the natural logarithm of the firm’s total assets; $Ret$ represents the annual stock return, expressed as the ratio of the year-end closing price to the annual opening price.

The residuals obtained from the model are used to measure the efficiency of the enterprise’s capital allocation, which in practice means the extent to which the enterprise’s actual level of new investment $Inv$ deviates from the expected optimal level of investment $\widehat{Inv}$, and $\epsilon$ represents the extent to which the enterprise allocates capital inefficiently. When $\epsilon$ is less than zero, it means that the firm is under-investing, and vice versa, it is over-investing. In other words, if $\epsilon$ deviates further from 0, it means that the enterprise’s capital allocation is less efficient. To test the model results more intuitively, the measure of enterprise capital allocation efficiency in this paper is denoted by −1000 times |$\epsilon$|, denoted as CAE, where the larger the CAE, the more efficient the enterprise capital allocation.

3.2.2. Explanatory Variables

The core explanatory variable in this paper is environmental information disclosure. In the current research, content analysis is commonly used to quantitatively analyze the quality of an enterprise’s environmental information disclosure, i.e., the textual content of an enterprise’s disclosure is quantitatively analyzed and ultimately presented in the form of data. This paper for the measurement system of environmental information disclosure indicators draws on studies of Clarkson (2008) [37] and Hussain et al. (2022) [38]. Considering the actual situation of environmental regulation in China in recent years, this paper designs a new evaluation system of environmental information disclosure indicators based on the studies mentioned above, and the specific evaluation system is shown in

Table 1. Environmental information disclosure system.

Group	Indicator	Grading Criteria
Enterprise management and governance structure	Environmental Management System	Develop a series of management systems such as relevant environmental management system and responsibility regulations, assign a value of 1, otherwise 0
	ISO14001 certification or not	Assign a value of 1 if the ISO14001 audit is passed, otherwise 0
	Annual Reports of Listed Enterprises	1 if the enterprise’s annual report discloses environment-related information, 0 otherwise
	Social Responsibility Report	CSR report disclosure of environment-related information is assigned a value of 1, otherwise 0
	Environmental Reports	Separate disclosure of environmental reports by enterprises is assigned a value of 1, otherwise 0
	Environmental honors or awards	Honors or awards received by the enterprise for environmental protection are assigned a value of 1, otherwise 0
	Three-Simultaneous System	If the “three-simultaneous” system is implemented, the value is 1, otherwise, it is 0
Environmental Performance Indicators	Pollutant emission compliance	If the pollutant discharge meets the standard, the value is 1, otherwise it is 0
	COD emissions	A total of 2 marks for quantitative description, 1 mark for qualitative description, and 0 marks for no description
	Wastewater emissions
	SO2 emissions
	CO2 emissions
	Fume and dust emissions
	Industrial solid waste generation
Current state of environmental governance	Exhaust emission reduction treatment	A total of 2 marks for quantitative description, 1 mark for qualitative description, and 0 marks for no description
	Wastewater abatement treatment
	Dust and fume control
	Solid waste utilization and disposal
	Noise light pollution radiation, etc. treatment
	Implementation of cleaner production
Spontaneous environmental behavior	Environmental Education and Training	Participation in environmental related education and training is assigned a value of 1, otherwise 0
	Environmental incident response mechanism	Establish an emergency response mechanism for major environment-related emergencies, assign a value of 1, otherwise 0
	Environmental Protection Special Action	A total of 1 for participation in environmental protection activities and social welfare activities, 0 otherwise
	Environmental philosophy	Establish environmental philosophy, environmental management structure, etc., assign a value of 1, otherwise 0
	Environmental objectives	Disclose the achievement of past environmental objectives and future environmental objectives, assign a value of 1, otherwise 0

.

3.2.3. Control Variables

In order to eliminate the influence of confounding factors on the causal identification results as far as possible, this paper further controls for six variables such as enterprise growth (Grow), the gearing ratio (Level), profitability (ROE), current ratio (Ratio), inventory turnover (Tur), and equity structure (GR) in the benchmark model based on Li et al. (2017) [39] and Busch and Hoffmann (2011) [40] on investment efficiency.

3.3. Model Construction

To examine the impact of environmental information disclosure on the efficiency of enterprise capital allocation, the following model is set up in this paper.

$$CA{E}_{i,t}={\beta }_0+{\beta }_{1}E{D}_{i,t}+Contro{l}_{i,t}\gamma +c_i+y_t+r_{pt}+{\sigma }_{i,t}.$$

(2)

In Equation (2), $CA{E}_{i,t}$ denotes the capital allocation efficiency of enterprise i in year t, measured using the Richardson model, ${\beta }_0$ is a constant term, $E{D}_{i,t}$ denotes the level of environmental information disclosure of enterprise i in year t, measured using the sum of the total value of the environmental information disclosure indicators established in this paper, and ${\beta }_1$ represents the estimated coefficient of the impact of environmental information disclosure on the capital allocation efficiency of enterprises. $Contro{l}_{i,t}$ represents a matrix composed of a series of control variables, $\gamma$ denotes the coefficient vector of this control variable matrix; $c_i$ denotes enterprise fixed effects, $y_t$ denotes year fixed effects, $r_{pt}$ denotes industry–time interaction fixed effects, and ${\sigma }_{i,t}$ denotes random disturbance terms. All variables’ symbols and specific meanings involved in Equation (2) are shown in

Table 2. Symbol of each variable and their definition.

Variables	Symbol	Definition
Capital allocation efficiency	CAE	1000 times the absolute opposite of the expected residuals of the investment model
Environmental Information Disclosure	ED	Overall score based on the environmental information disclosure indicator system
Financing constraints	SA	Constructing an SA index model by drawing on Hadlock and Pierce’s study
Enterprise growth	Grow	(Operating income for the year—operating income for the previous year)/operating income for the previous year
Gearing ratio	Level	Total liabilities/total assets
Profitability	ROE	Net profit/average balance of shareholders’ equity
Liquidity ratio	Ratio	Current assets/Current liabilities
Inventory turnover rate	Tur	Cost of goods sold/Closing balance of inventories
Shareholding structure	GR	Percentage of shareholding of controlling shareholders

.

4. Results

4.1. Descriptive Statistics and Correlation Analysis

This section begins with a descriptive statistical analysis of the variables, the corresponding results are shown in

Table 3. Descriptive statistics of variables.

Variable	N	Mean	Medium	Sd	Max	Min
CAE	4230	−64.470	−45.970	78.690	−0.028	−1219.000
ED	4230	10.630	9.000	7.693	35.000	1.000
Grow	4230	0.138	0.063	0.914	43.910	−0.969
Level	4230	0.438	0.436	0.215	1.655	0.014
ROE	4230	0.032	0.039	0.281	6.513	−5.745
Ratio	4230	2.143	1.403	3.016	84.930	0.152
Tur	4230	18.830	10.920	26.910	201.100	0.910
GR	4230	35.950	34.090	15.960	89.990	0

. It can be seen that the maximum value of ED is 35, the minimum value is 1, the mean value is 10.63, and the standard deviation is 7.693, indicating that the quality of environmental disclosure varies among listed enterprises in the heavily polluting industries. The maximum value of enterprise CAE is −0.028, the minimum value is −45.97, the mean value is −64.47, and the standard deviation is 78.69, indicating significant differences in the capital allocation efficiency of the enterprises in the study sample.

For the statistical aspects of the control variables, the high mean values of growth and profitability indicate that most of the enterprises in the sample have good economic efficiency in their own investments, are less risky for investors and are worthy of continued investment, and have a certain degree of growth, but there are some loss-making enterprises as can be seen from the minimum values. The mean values of the three financial indicators of gearing ratio, current ratio, and inventory turnover ratio remained within the normal indicator range, indicating that most of the sample enterprises were operating well. The above indicator data also reflects the better performance of heavy polluters in the capital market and the higher potential investment opportunities.

4.2. The Impact of ED on CAE of Enterprises

To test hypothesis H1, model (2) is estimated using the OLS method, and the estimation results are presented in

Table 4. Regression results of environmental information disclosure and efficiency of enterprise capital allocation.

Variable	(1)	(2)	(3)	(4)	(5)
	CAE	CAE	CAE	CAE	CAE
ED	0.790 ***	0.747 ***	1.575 ***	0.926 ***	0.762 **
	(0.143)	(0.155)	(0.291)	(0.323)	(0.325)
Grow		−2.871	−0.246	−0.803	−0.563
		(2.324)	(2.562)	(2.584)	(2.438)
Level		0.884	24.453	26.196	18.482
		(8.683)	(21.639)	(21.564)	(22.418)
ROE		10.451	6.439	6.234	6.631
		(7.547)	(8.743)	(8.997)	(9.210)
Ratio		−0.281	0.122	0.520	0.809
		(0.422)	(0.528)	(0.520)	(0.556)
Tur		0.041	0.300 ***	0.273 ***	0.234 **
		(0.051)	(0.090)	(0.091)	(0.093)
GR		0.030	−0.088	0.248	0.227
		(0.072)	(0.214)	(0.219)	(0.221)
Constant	−72.865 ***	−73.962 ***	−94.846 ***	−101.081 ***	−95.129 ***
	(2.238)	(4.443)	(14.381)	(14.305)	(14.751)
Enterprise fixed effect	N	N	Y	Y	Y
Year fixed effects	N	N	N	Y	Y
Year Industry fixed effects	N	N	N	N	Y
N	4230	4230	4228	4228	4225
R2	0.006	0.009	0.270	0.278	0.304

Note: 1. ** and *** in the tables indicate significant at the 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. Column (1) does not control for year-fixed effects, enterprise-fixed effects, year industry fixed effects, and other control variables, the estimation results in this column indicate that ED has a significant positive promotion effect on enterprises’ CAE; however, this result has a large omitted variable bias. Therefore, columns (2), (3), (4), and (5) control for other control variables, enterprise fixed effects, year fixed effects, and year industry fixed effects gradually, and the estimated coefficients do not fluctuate significantly and remain significantly positive at the 5% level. This indicates that ED has a significant positive effect on enterprises’ CAE, which means that enterprises can significantly improve their CAE through ED. Hypothesis H1a is proved.

4.3. Robustness Tests

4.3.1. Replacement of ED’s Measure

The direct summation method used in this paper to measure ED ignores the correlation between the indicators and may lead to measurement errors in the indicators, therefore, this paper refers to the existing research [41] and recalculates the environmental information disclosure indicator (ED1) using the entropy weighting method, and then uses model (2) to test the results, which are shown in column (1) of

Table 5. Robustness test results.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	CAE	CAE	CAE	CAE	CAE	CAE
ED1	4.395 **
	(2.096)
ED		0.762 **	0.966 ***	0.995 ***	0.762 **	4.487 ***
		(0.382)	(0.319)	(0.319)	(0.325)	(0.703)
p1			6.823 *
			(4.073)
p2				−7.397 *
				(3.818)
Constant	−87.407 ***	−95.129 ***	−4400.446 **	−8973.154 ***	−95.129 ***
	(13.472)	(16.885)	(1709.345)	(1949.530)	(14.751)
Kleibergen-Paap rk LM						532.236 ***
Kleibergen-Paap rk Wald F						383.008
						(19.93)
Hansen J						0.379
Control variables	Y	Y	Y	Y	Y	Y
Enterprise fixed effect	Y	Y	Y	Y	Y	Y
Year fixed effects	Y	Y	Y	Y	Y	Y
Year Industry fixed effects	Y	Y	Y	Y	Y
N	4225	4225	4228	4228	4225	4221
R2	0.304	0.304	0.277	0.277	0.304	−0.017

1. *, ** and *** in the tables indicate significant at the 10%, 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. The results indicate that after replacing the measures of environmental information disclosure, it still significantly promotes the CAE of enterprises. This result is consistent with the results in Table 4.

4.3.2. Cluster-Robust Standard Error

Considering many enterprises may be distributed within the same industry, disturbances for enterprises within the same industry may be correlated. This may lead to an underestimation of the standard errors of the estimated coefficients. To address this issue, we cluster the standard errors at the industry level for robustness tests based on model (2). The estimation results are reported in column (2) of Table 5. It can be seen that the estimated coefficient of ED is 0.762 and significant at the 5% level, which is consistent with the previous conclusion.

4.3.3. Excluding Other Policies

Although the reliability of the regression results is further guaranteed by the robustness tests above, the implementation of any policy will inevitably be affected by other policies or historical shocks, which, in turn, will have an impact on the assessment of the effects of the policy in question. Therefore, this paper excludes shocks from other policies in this section, specifically: first, to eliminate the impact of the new environmental protection law that came into effect in January 2015 and the Construction of the National Carbon Emissions Trading Market (Power Generation Sector) that went into effect in 2018 on the estimated results of this paper, this paper adds the dummy variable p1 for the new environmental protection law and the policy dummy variable p2 for the “Construction of National Carbon Emission Trading Market (Power Generation Sector)” to the regression equation. The corresponding estimation results are reported in columns (3) and (4) of Table 5, respectively. Secondly, the current policy on the disclosure of environmental information by enterprises in China has made it mandatory for critical emission units to disclose their environmental information. In order to exclude the influence of institutional factors on enterprises’ choice of disclosure vehicle [42], critical emission units are excluded from the regressions, and the corresponding results are reported in column (5) of Table 5. The test results in columns (3), (4), and (5) of Table 5 show that the coefficient on ED remains significant after controlling for relevant impact policies and excluding critical emission units.

4.3.4. Instrumental Variables

Since there may be a reverse causality between ED and CAE, which may interfere with the estimation results of this paper. To address this issue, referring to Hasan et al.’s (2015) study, the average of all companies’ ED in the same industry and within the same year was used as an instrumental variable, the reason for choosing this variable as an instrumental variable is that there is a “peer effect” [43] in the disclosure of environmental information by firms in the same industry. When the quality of environmental information disclosure in the industry is higher, the quality of environmental information disclosure by enterprises in the industry is likely to be higher, which satisfies the relevance of the instrumental variable. In addition, the exogeneity assumption of the instrumental variable is satisfied as the quality of environmental information disclosure for the industry as a whole is an industry variable and cannot be determined by the decision-making behavior of a single enterprise.

Furthermore, this paper chose whether the auditors of the sample companies were from the “Big Four” firms as the instrumental variable. The reason for this is that if a listed company has an auditor from one of the “Big Four” firms, the quality of disclosure is generally higher, thus satisfying the correlation of the instrumental variable, while the capital allocation efficiency of the enterprise is not directly related to the type of auditor firm, thus fulfilling the exogeneity hypothesis of the instrumental variable [44].

The regression results are reported in column (6) of Table 5. From this, the Kleibergen-Paap RK LM statistic is 532.236 and significant at the 1% significance level, indicating no unidentifiability of instrumental variables. The value of the Kleibergen-Paap rk Wald F statistic is 383.008 and much greater than the value of 19.93 at the 10% critical value level, indicating that there is no weak instrumental variable problem. Hansen J’s p-value of 0.379 suggests that the original hypothesis that the instrumental variables are exogenous cannot be rejected. The estimation results show that the estimated coefficient of ED is positive and significant at the 1% level, indicating that ED promotes CAE, which is consistent with the findings from the benchmark regression.

4.4. Heterogeneity Analysis

4.4.1. Enterprise Size

In general, economies of scale usually put larger enterprises in an advantageous position regarding CAE. Conversely, small-scale enterprises are constrained by their conditions to improve ED while also increasing the uncertainty of operational efficiency, which may reduce their CAE. To test the above perspective, this paper divides the total sample into two categories: large-scale and medium- and small-scale enterprises.

Table 6. Heterogeneity analysis of enterprises’ size.

Variable	(1)	(2)
	Large Enterprises	Small and Medium-Sized Enterprises
ED	0.073	1.708 **
	(0.317)	(0.717)
Constant	−98.231 ***	−93.025 ***
	(17.519)	(24.196)
Control variables	Y	Y
Enterprise fixed effect	Y	Y
Year fixed effects	Y	Y
Year Industry fixed effects	Y	Y
N	2083	2060
R2	0.389	0.330

1. ** and *** in the tables indicate significant at the 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

reports the results of grouped regressions for enterprises of different sizes, with column (1) showing the estimated results for large-scale enterprises and column (2) for medium- and small-scale enterprises. The estimated coefficient of ED is positive and significant at the 5% significance level in the regressions for large-scale firms. In contrast, the effect of ED on the CAE of enterprises is not significant for medium- and small-scale enterprises. Indicating that the impact of ED on the CAE of enterprise varies according to the size of the enterprise. Hypothesis H2 is proved.

4.4.2. Enterprise’s Ownership

The analysis in the hypothesis section of the study indicates that the impact of ED on the CAE of an enterprise varies according to the ownership of the enterprise. To test whether this phenomenon exists, the study sample was divided into two sub-samples of state-owned (SoEs) and non-state-owned enterprises (non-SoEs) based on enterprise ownership, and the results are reported in

Table 7. Heterogeneity analysis of enterprise ownership.

Variable	(1)	(2)
	State-Owned Enterprises	Non-State Enterprises
ED	1.067	0.621 *
	(1.025)	(0.354)
Constant	−121.609 **	−92.972 ***
	(50.941)	(16.774)
Control variables	Y	Y
Enterprise fixed effect	Y	Y
Year fixed effects	Y	Y
Year Industry fixed effects	Y	Y
N	469	3703
R2	0.615	0.309

1. *, ** and *** in the tables indicate significant at the 10%, 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. It shows that there are differences in the impact of ED on the CAE of an enterprise for its different ownership. In the sub-sample of non-SOEs, the estimated coefficient of ED is 0.621 and is significant at the 10% level. In contrast, the estimated coefficient of ED in the sub-sample of non-SoEs is 1.067 but not significantly different from zero. Therefore, Hypothesis H3 is proved.

4.4.3. Enterprise Life Cycle

According to life cycle theory, the enterprises’ business strategies, growth, profitability, and willingness to disclose the environment differ significantly at different stages of development, resulting in differences in the severity of information asymmetry and agency problems they face, which inevitably affects the efficiency of capital allocation. Therefore, the effect of ED on the CAE of the enterprise may be different depending on its life cycle stages.

In order to test whether the above phenomenon exists, this paper refers to the cash flow model method proposed by Dickinson (2011) [45] and classifies the life cycle of enterprises into three stages of “growth—maturity—recession” and uses the benchmark regression model to test the grouping of enterprises with different life cycles, the results are shown in

Table 8. Heterogeneity analysis of enterprises’ lifecycle.

Variable	(1)	(2)	(3)
	Growth Stage	Maturity Stage	Decline Stage
ED	1.040 *	0.060	−1.057
	(0.590)	(0.338)	(2.143)
Constant	−143.774 ***	−64.477 ***	47.569
	(35.586)	(13.237)	(47.060)
Control variables	Y	Y	Y
Enterprise fixed effect	Y	Y	Y
Year fixed effects	Y	Y	Y
Year Industry fixed effects	Y	Y	Y
N	1689	1645	476
R2	0.492	0.399	0.639

1. * and *** in the tables indicate significant at the 10% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. In column (1), the estimated coefficient of ED in growing enterprises is 1.04. It is significantly positive at the 10% level, indicating that mature enterprises can contribute to CAE through ED. The estimated coefficients of ED for mature and declining enterprises are 0.06 and −1.057 but they are not significant. Therefore, Hypothesis H4 is proved.

4.4.4. Region Where the Enterprise Is Located

Considering that factors such as the level of economic development in an enterprise’s region may impact its production and operation activities, such impact may lead to regional heterogeneity in the effect of ED on enterprises’ CAE. In order to test whether such heterogeneity exists, this paper divides the enterprises into four sub-samples: eastern, central, western, and northeastern. Model (2) was used to test the impact of ED on the CAE of corporate in different sub-samples. The results are reported in

Table 9. Heterogeneity analysis of enterprises’ location.

Variable	(1)	(2)	(3)	(4)
	Eastern Region	Central Region	Western Region	Northeast Region
ED	0.868 *	1.412 ***	−1.290	0.885
	(0.453)	(0.513)	(0.917)	(1.552)
Constant	−106.790 ***	−78.644 ***	−17.889	−180.780 **
	(22.665)	(15.699)	(38.552)	(81.498)
Control variables	Y	Y	Y
Enterprise fixed effect	Y	Y	Y	Y
Year fixed effects	Y	Y	Y	Y
Year Industry fixed effects	Y	Y	Y	Y
N	2334	836	790	179
R2	0.298	0.366	0.563	0.651

1. *, ** and *** in the tables indicate significant at the 10%, 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. The results show that ED has a significant positive contribution to the CAE of enterprises in the Eastern and Central regions. In contrast, in the Western and Northeastern regions, the coefficient of ED is not significant. Therefore, Hypothesis H5 is proved.

4.5. Stakeholders’ Moderating Role

The research hypothesis section suggests that there may be a moderating role for enterprise stakeholders in the process of ED impacts on the CAE of enterprises and to test whether this role exists, the following moderating effect model is constructed in this section for analysis.

$$CA{E}_{i,t}={\beta }_0+{\beta }_{1}E{D}_{i,t}+{\beta }_{2}ED\times ES+{\beta }_{3}ES+Contro{l}_{i,t}\gamma +c_i+y_t+r_{pt}+{\sigma }_{i,t},$$

(3)

$$CA{E}_{i,t}={\beta }_0+{\beta }_{1}E{D}_{i,t}+{\beta }_{2}ED\times OR+{\beta }_{3}OR+Contro{l}_{i,t}\gamma +c_i+y_t+r_{pt}+{\sigma }_{i,t},$$

(4)

$$CA{E}_{i,t}={\beta }_0+{\beta }_{1}E{D}_{i,t}+{\beta }_{2}ED\times COF+{\beta }_{3}COF+Contro{l}_{i,t}\gamma +c_i+y_t+r_{pt}+{\sigma }_{i,t}.$$

(5)

In Equations (3)–(5), ES, OR, and COF are the proxy variables for stakeholder satisfaction of the enterprise’s employees, shareholders, and creditors, respectively. Where the natural logarithm of employee compensation payable is used as a measure of stakeholder satisfaction of the company’s employees. The natural logarithm of owner’s equity is a measure of shareholder interest satisfaction, and enterprise finance costs as a measure of stakeholder satisfaction of enterprise creditors [46]. The meanings of the other symbols in the three models above are consistent with those in Equation (2), and the corresponding estimation results for each model are reported in

Table 10. Tests for moderating effects of stakeholders.

Variable	(1)	(2)	(3)
	Employee	Shareholder	Creditors
ED	8.115 ***	9.262 **	1.707 ***
	(2.846)	(3.747)	(0.302)
ED*ES	−0.415 ***
	(0.154)
ES	5.762 **
	(2.779)
ED*OR		−0.390 **
		(0.159)
OR		15.706 ***
		(3.438)
ED*COF			−4 × 10−10 ***
			(1 × 10−10)
COF			1.6 × 10−8 ***
			(3.8 × 10−9)
Constant	−195.383 ***	−423.097 ***	−96.914 ***
	(52.204)	(76.785)	(14.433)
Enterprise fixed effect	Y	Y	Y
Year fixed effects	Y	Y	Y
Year Industry fixed effects	Y	Y	Y
N	4188	4205	4217
R2	0.306	0.032	0.272

1. ** and *** in the tables indicate significant at the 5% and 1% levels, respectively; 2. Robust standard errors are in brackets.

. The estimated coefficient of the interaction term ED*ES is negative at the 1% significance level, indicating that employee satisfaction has a negative moderating effect in the process of ED acting on the CAE of the enterprise. The estimated coefficient of the interaction term ED*OR is negative and significant at the 5% significance level, indicating that the moderating effect of shareholder satisfaction is not observed in the impact of ED on the CAE of the enterprise. The estimated coefficient of the interaction term ED*COF is negative and significant at the 1% significance level, indicating a negative moderating effect of creditor satisfaction in the process of ED on the CAE of enterprise. Therefore, Hypothesis H6 is proved.

5. Conclusions and Discussion

5.1. Research Conclusions

This paper uses panel data from heavily polluting industries in A-share listed companies in China from 2013 to 2020 and constructs econometric models to analyze the impact of environmental information disclosure on the efficiency of capital allocation in enterprises from both theoretical and empirical perspectives. The main conclusions are as follows:

(1)

There is a significant and robust promotion effect of environmental information disclosure on the capital allocation efficiency of enterprises. Based on the panel data of enterprises in the heavy pollution industry among A-share listed enterprises in China from 2013 to 2020, after analyzing the impact of environmental information disclosure on the capital allocation efficiency of enterprises using a fixed-effects model, it is found that environmental information disclosure can significantly promote the capital allocation efficiency of enterprises. The findings are consistent with the benchmark regression after robustness tests are conducted in four ways: changing the indicator measure, clustering robust standard errors, excluding another policy confounding, and selecting instrumental variables. This result is consistent with the hypothesis of legitimacy theory, which suggests that the disclosure of environmental information influences the perception of legitimacy by stakeholders, reduces information asymmetry within and outside the enterprise, helps external investors understand the enterprise’s operating conditions, reduces inefficient investment behavior, and improves the efficiency of the enterprise’s capital allocation.

(2)

Based on the characteristics of enterprise size, ownership, life cycle stage, and region, this study divided the sample into different subgroups for regression analysis. Firstly, environmental information disclosure has a significantly positive impact on the capital allocation efficiency of small and medium-sized enterprises at a significance level of 5%, but this phenomenon was not observed in large-scale enterprise samples. Secondly, environmental information disclosure has a significantly positive impact on the capital allocation efficiency of non-state-owned enterprises at a significance level of 10%, but this effect was not observed in state-owned enterprise samples. Thirdly, environmental information disclosure has a significantly positive impact on the capital allocation efficiency of enterprises in the growth stage at a significance level of 10%, but this effect was not observed in samples of enterprises in the mature and decline stages. Fourthly, environmental information disclosure has a significant positive effect on the capital allocation efficiency of enterprises in the eastern and central regions, but no significant impact was observed in the western and northeastern regions.

(3)

Stakeholders play a negative moderating role in the process of environmental information disclosure affecting the efficiency of enterprise capital allocation. Moderating effect models were constructed and found that employees, shareholders, and creditors all have a negative moderating effect in the process of environmental information disclosure affecting the efficiency of corporate capital allocation. The reason for this phenomenon is that if an enterprise’s environmental disclosure is of high quality, the enterprise needs to invest more resources in environmental activities, which can crowd out the rights of these stakeholders and lead to resistance from them.

5.2. Implications

According to the research findings, this paper proposes the following recommendations to improve the efficiency of capital allocation in enterprises:

(1)

Enhancing the intrinsic motivation for environmental information disclosure. Environmental information disclosure is one of the important tools for enterprises to carry out environmental and social responsibility governance. In the current context where environmental protection is receiving increasing attention from the government and the public, although choosing to disclose environmental information will generate certain economic costs and burdens for the company in the short term, in the long term, corporate managers should actively disclose environmental information to the public, incorporate environmental information disclosure into strategic management, set reasonable environmental performance targets, and enhance employees’ environmental awareness, establish and improve the company’s environmental management system and processes, thereby enhancing the company’s environmental responsibility and consolidating its legitimacy.

(2)

Promoting the heterogeneous development of environmental information disclosure in enterprises. Because the impact of environmental information disclosure on the efficiency of capital allocation in enterprises is heterogeneous in terms of company size, property rights, life cycle, and region, companies should set environmental goals that are consistent with their own characteristics based on their different stages of development. For large-scale enterprises, they should further improve their asset management processes and enhance resource allocation efficiency. For state-owned enterprises, they should be more proactive in disclosing environmental information. For mature and declining enterprises, in addition to focusing on their own development issues, they should also pay more attention to the environmental impact of their activities and disclose relevant information. For enterprises in western and northeastern regions, they should strengthen exchanges and learning with enterprises in eastern and central regions.

(3)

Coordinating the rights of various stakeholders in a reasonable manner. Stakeholders have interests that are closely tied to the realization of the company’s goals. However, different stakeholders have different interests, which can lead to conflicts, and their demands can also change at different stages of the company’s development. Therefore, in the process of governing the company, corporate managers should further coordinate the interests of stakeholders such as employees, shareholders, and creditors to achieve a balance of interests. Specifically, for employees, managers should delegate some power to compensate for the resources that may be squeezed due to environmental information disclosure, such as improving employee compensation and job grades. For shareholders, companies should maintain good relationships with them, regularly report on the company’s operating status, and reduce the problem of information asymmetry between shareholders and corporate managers. For creditors, companies should timely disclose the company’s capital operation situation and do a good job in project risk management to ensure the safety and returns of creditors’ funds.

5.3. Discussion

In recent years, the relationship between ED and firms’ economic performance has attracted more and more scholars’ attention. CAE also reflects the economic performance of firms, but few scholars have paid attention to the potential impact of ED on the CAE of firms, many studies instead research the impact of ED on enterprises’ Tobin’s Q, ROA, and stock prices [11,12,13,14,15]. In this paper, we further focus on the impact of ED on corporate CAE based on the existing studies and find ED has a significant promotion effect on corporate CAE, and this result is consistent with the previous studies, such as the result of Chouaibi et al. [47], Emuebie et al. [48], and Alipour et al. [49], that is, ED is significantly positively correlated with corporate economic performance. Actually, these authors find that board independence and the fulfillment of social and ethical responsibilities by firms strengthen the promoting effect of ED on firm performance. However, this paper finds that stakeholder pressure weakens the promoting effect of ED on firms’ CAE, this finding, while different from previous studies, is consistent with stakeholder theory. The reason for this phenomenon is that firms’ performance is always affected by the external environment and their relationships with stakeholders, and that higher quality of firms’ ED means that firms have invested more resources in environmental protection activities, which will crowd out the resources used by the firm to enhance the treatment of its employees, shareholders, and creditors, leading to boycotting behavior by these stakeholders. However, limitations also exist. Firstly, to eliminate the potential bias caused by industry characteristics on estimation results, this study focuses solely on companies in heavily polluting industries. This may lead to sample selection bias. Furthermore, as the empirical analysis results have only internal validity, the impact of environmental information disclosure on capital allocation efficiency in non-heavy-polluting industry companies remains unknown. Furthermore, to measure environmental information disclosure, this study assigns values by designing an indicator system. However, a company’s environmental information disclosure involves a wide range of content, and the indicator system designed in this study may not have included all the main content of environmental information disclosure. In addition, this study only analyzes the impact of environmental information disclosure on capital allocation efficiency at the micro-level of companies, without considering whether there is a similar effect at the industry-level and regional macro-level.

Therefore, there are possibilities for further deepening this research. On the one hand, as low-carbon and green concepts gain traction, environmental issues caused by corporate production are increasingly receiving public attention. This puts forward new requirements for corporate environmental information disclosure. The existing evaluation indicators and measurement methods for corporate environmental information disclosure quality should adapt to the development of the times and establish a more scientific, reasonable, and comprehensive indicator system. Further scientific measurement methods should also be developed to guide companies toward a resource-efficient production mode. On the other hand, non-heavy-polluting industry companies are also affected by the continuous introduction and implementation of environmental governance policies and have disclosed corresponding environmental information. How environmental information disclosure affects their capital allocation efficiency is also a question worth considering.