Does Public Environmental Attention Improve Green Investment Efficiency?—Based on the Perspective of Environmental Regulation and Environmental Responsibility

Abstract

Will public attention to the environment affect the green behavior of enterprises? This paper collected the panel data of steel enterprises in China from 2009 to 2017, evaluated the green investment efficiency by the measurement of the slack model (SBM) and the “super efficiency” (DEA) model, and studied the impact of public environmental attention on green investment efficiency. Firstly, public environmental attention has significantly promoted enterprise green investment efficiency; Secondly, the business environment has significantly promoted enterprise green investment efficiency, and the influence of public environmental attention on enterprise green investment efficiency is positively regulated by the business environment; Thirdly, the influence mechanism of public environmental attention on the promotion of green investment efficiency comprises environmental regulation and environmental responsibility; Fourthly, public environmental attention has a heterogeneous effect on the promotion of enterprise green investment efficiency. The promotion of public environmental attention on green investment efficiency is more significant for large enterprises than medium-sized enterprises. The effect of the promotion of public environmental attention on green investment efficiency is more significant for state-owned enterprises than private enterprises. The promotion of public environmental attention on green investment efficiency is more significant in eastern China than in central and western China.

1. Introduction

Since the start of the 21st century, China’s economy has achieved remarkable results [1]. However, due to its neglect of environmental management, economic growth has also brought a series of serious environmental pollution problems [2]. In 2019, the bulletin on the state of China’s ecological environment disclosed that there were 180 cities (more than half of the total number of cities) where the air quality seriously exceeded the standard, and the total time of serious pollution in domestic cities is up to 1666 days—an increase of 88 days over 2018. Therefore, it is imperative to strengthen China’s environmental governance.

With the rapid development of the Internet, public attention has gradually evolved into an important new form of environmental management, especially with regard to the ecological environment [3]. The high public demand for good environmental quality indicates the urgency with which the public desire environmental improvements [4,5]. In order to maintain social stability, the central government collects public environmental appeals through various channels, summarizes various environmental problems through classification, and formulates corresponding environmental pollution laws and regulations. With the increasing support of the government for environmental issues, the public’s attention to the environment is increasingly irreplaceable in social environmental management [6]. Research has shown that the public’s attention to the environment not only promotes more active public participation in environmental management [7], but can also deal with the specific problems involved in environmental pollution management [8]. The public’s perception of environmental protection issues can not only improve environmental awareness, but also help to find and solve social conflicts [9], which is beneficial to China’s economy [7]. Therefore, public environmental attention is likely to evolve into an important means of improving the social environment in China, while the lack of attention may relax the government’s environmental protection supervision [10].

Enterprises are the main body of the national economy. While promoting economic growth, enterprises also consume a lot of energy, causing significant environmental pollution [11]. According to environmental statistics data, sulfur dioxide, nitrogen oxides, and industrial smog emissions of enterprises across the country remain high and are seriously damaging China’s air quality. Therefore, the Chinese government encourages enterprises to improve their environment through green transformation. In recent years, the emergence of green investment has attracted many enterprises. More and more enterprises are trying to improve their environment through green investment [12]. However, at this stage, many enterprises in China are at the trial stage of green investment, and the problem of an insufficient degree of green investment is very common. The development level of green investment is low, green governance is relatively backward, and the effect of green investment is not ideal [13,14]. China’s environmental pollution problem has not been resolved, which also hinders high-quality development. Therefore, reasonable support for enterprise green investment is a good means of improving environment.

This paper makes several contributions. Firstly, this paper studies enterprise green investment through public attention. Previous research has mostly focused on the contribution of emission reduction [15], as well as various internal factors affecting green investment, such as enterprise size and industry heterogeneity [16]. Through the influence of public attention, the improvement in enterprise green investment efficiency is conducive to improving sustainable development [17].

Secondly, the frequency of environmental protection words according to the Baidu search index was used to measure the public environmental attention. In previous studies, public environmental attention and participation were mostly measured by official statistical data such as environmental letters, letter reviews, and the proposals of the CPPCC. Such data generally have shortcomings such as low efficiency, high cost, and incomplete information response. In recent years, the Internet search index has gradually been promoted and used [18,19]. The Baidu search index was selected to estimate public attention to environmental protection, and provides a reference for enriching the research on public environmental attention.

Finally, this paper studies the influence mechanism between the two variables. Most of the previous research has focused on public environmental attention to air pollution [20]. This article combs the influence of public environmental attention at the micro-enterprise level through the two elements of “environmental regulation” and “ environmental responsibility”, and further examines the influence on green investment efficiency. In addition, this paper also carries out the corresponding heterogeneity research on the enterprise scale, regional environment, and other aspects. The framework is shown in Figure 1.

2. Theoretical Analysis and Research Hypothesis

In recent years, environmental protection has received more and more public attention. Public attention to the environment indicates the public’s concern over the current environmental situation and expectations of a better environment in the future [4]. China is a centralized country. When environmental problems cause a series of public and social problems, the government will conduct constraint management on enterprises [5]. Public environmental pressure will urge local governments to strengthen the implementation of corresponding environmental regulations and provide more environmental services [21]. Under the pressure of the public and governments at all levels, enterprises are forced to increase their input in environmental governance to improve the enterprise environment [22]. Firstly, the public’s attention and demands in terms of environmental issues will attract the attention of the government, and the government will implement the stricter supervision of enterprises [23]; Secondly, in order to avoid excessive environmental penalties, enterprises must increase environmental protection measures to improve the green investment efficiency [24]; Finally, although green investment brings uncertainty to the income of enterprises, it cannot make enterprises derive economic profits in the short term. However, through green investment, the energy utilization rate of enterprises can be greatly improved, which also promotes their economic benefits to a certain extent [25,26].Based on the above analysis, we can obtain the following hypothesis:

Hypothesis 1 (H1).

Public environmental attention has a significant role in promoting enterprise green investment efficiency.

A business environment is the basic condition for enterprise development. On the one hand, a good business environment not only requires enterprises to abide by environmental protection rules, jointly maintaining and creating a better business environment, but also improves the return of the natural resources of enterprises and green investment efficiency by standardizing enterprise behavior [27]. On the other hand, research has shown that the business environment has significantly improved enterprise performance [28]. A good business environment not only makes it easier for enterprises to make profits [29], but also makes a contribution to social responsibility, making enterprises pay more attention to their own green reputation on the premise of making profits, and improve the status of enterprises in the hearts of consumers by paying attention to environmental protection [30]. When the public pays significant attention to environmental issues, enterprises have the strength and willingness to expand green investment in a good business environment. Based on the above analysis, we can formulate the following hypothesis:

Hypothesis 2 (H2).

The business environment plays a significant role in promoting green investment efficiency. The impact of pubic environmental attention on green investment efficiency is positively moderated by the business environment.

The research shows that when a social problem has attracted enough public attention, it becomes more urgent to resolve such problems [31,32]. For the government, when environmental issues do not receive enough attention from the public, the issue of environmental pollution will naturally not attract attention. It is only when the problem becomes excessively serious to be tolerated by the public and may cause relevant social conflicts that the government must take a series of measures to solve environmental problems in the face of public pressure, such as by strengthening environmental supervision, strictly implementing environmental laws and regulations, and incorporating environmental protection indicators into the assessment systems of government officials, in order to effectively regulate the behavior of enterprises and implement environmental protection practices [33]. From this perspective, public concern over environmental pollution has led to stricter environmental policies [34]. A study found that the strengthening of environmental regulation directly leads to the withdrawal and transfer of some local polluting enterprises [35], while remaining enterprises staying will pay more attention to investment in green environmental protection. Research has shown that environmental regulation improves the entry threshold of polluting enterprises in the region, and has a filtering effect on pollution industries [36]. Environmental regulation will reduce the entry rate of polluting enterprises and improve the scale and level of green investment in the region to a certain extent [37]. For enterprises, strict environmental regulations positively contribute to reducing pollution emissions and improving the environmental performance of enterprises [38]. However, green investment requires lots of capital and technological investment, for which the funds of many enterprises—particularly those in financial distress—are insufficient. Green investment not only increases the operating risk of enterprises [39], but also seriously reduces the willingness of enterprises to make green investments [40]. Therefore, when environmental regulation is relatively loose, some enterprise green investments are often made to reduce environmental penalties and meet the management requirements of the government, but the efficiency of such green investment is very low [41]. When environmental regulation is continuously strengthened (i.e., the critical value and inflection point), the amount of green investment required is far lower than environmental taxes. At this time, enterprises can no longer avoid their own pollution control problems. Especially when the public pays attention to environmental problems or polluting enterprises themselves, enterprises will enhance their green investment efficiency and environmental governance effect, and closely abide by environmental policies [15]. Combined with the above analysis, we formulate the following hypothesis:

Hypothesis 3 (H3).

Public environmental attention takes environmental regulation as the mediating variable to promote green investment efficiency.

Green investment will increase the cost of enterprises to a certain extent, which is not conducive to improving enterprise competitiveness [41]. However, through green investment, enterprises can show the public their willingness to take green action and display their eagerness actively undertake environmental responsibility. This behavior can not only improve the green reputation of the enterprise, but also supervise the environmental protection practice of the enterprise to a certain extent, and then improve the efficiency of green investment [42,43].

In the context where the public pays great attention to the environment, enterprises can effectively improve the influence and competitiveness of their products in the market by maintaining the green reputation of their brands through green practices [44]. In addition, based on the social responsibility of environmental protection, enterprises meet the environmental protection needs of customers, governments, and other stakeholders through green investment, so as to obtain the support of stakeholders to improve enterprise performance [45]. Research shows that, compared with low-environmental-responsibility enterprises, high-environmental-responsibility enterprises tend to obtain higher long-term profitability through green investment [46]. Therefore, through environmental responsibility, public environmental attention can effectively improve the willingness of green investment, thereby improving the efficiency of green investment. Combined with the above analysis, we formulate the following hypothesis:

Hypothesis 4 (H4).

The public environmental attention takes environmental responsibility as the mediating variable to promote green investment efficiency.

3. Data, Methodology, and Model

3.1. Data Selection of Green Investment Efficiency

As China’s pillar industry, the steel industry is responsible for significant environmental pollution. Therefore, the production of steel may attract great attention from the Chinese people. Due to the industry’s characteristic of high emissions and heavy pollution, this paper selected China’s large- and medium-sized steel enterprises whilst considering enterprise green investment as input; the waste produced by the enterprise as the unexpected output; and the output value of the comprehensive utilization of three wastes as the expected output to measure the green investment efficiency of enterprises. The panel data of the 50 large enterprises involved in this paper from 2009 to 2017 were obtained from China Iron and Steel Association. The explanation and statistical description of data variables are shown in

Table 1. Variable selection and interpretation.

Variables	Name	Interpretation
Input	Green investment	Total annual funds used by the enterprise for purchasing environmental protection equipment, R&D of environmental protection technology, and pollutant treatment
Undesired outputs	Waste residue	Total amount of waste residue produced by enterprises
	Waste gas	Total amount of waste gas produced by enterprises
	Waste water	Total amount of waste water produced by enterprises
Expected output	Output value of three wastes utilization	Refers to the value of products produced using the “three wastes” (waste water, waste gas, and waste residue) as the main raw materials

and

Table 2. Descriptive statistics of the input and output variables for steel enterprises (2009–2017).

Variables	Units	Mean	Std. Dev	Min	Max
Green investment	Ten thousand yuan	19,588.85	34,963.52	570	268,939
Waste residue	Ten thousand tons	431.26	373.15	1.98	2353.52
Waste gas	Hundred million ${\mathrm{m}}^3$	1492.31	1145.36	0.17	7353.02
Waste water	Million ${\mathrm{m}}^3$	566.61	474.42	1.28	3450.80
Output value of three wastes utilization	Ten thousand yuan	61,094	86,057.83	275	595,652

, respectively.

3.2. Measuring Enterprise Green Investment Efficiency

DEA is a widely used efficiency measurement method. The DEA method has great advantages over a parameter estimation method, especially in the evaluation of enterprise efficiency. When doing micro-enterprise research, the use of parameter estimation in this case will lead to a large deviation in the input–output matching parameters of different enterprises, and it is impossible to use a unified standard to measure enterprise performance. The DEA method only needs to consider the input and output units, not even the unit of data, so as to avoid the error in parameter estimation. Therefore, in the research on micro-enterprises, the DEA method is more commonly used than parameter estimation. According to the relevant research on the quantification of the comprehensive performance of highly polluting enterprises [47], the integration of resources and environmental factors into the comprehensive performance evaluation of enterprises has always been the focus of academic attention. Due the steel industry’s characteristic high levels of consumption, emissions, and pollution, research on steel enterprises must fully consider their undesirable outputs, including pollutant emission. Therefore, the ordinary DEA efficiency model will no longer be applicable to such research, and the DEA model with undesired output needs to be introduced to deal with such problems.

Based on the defects and limitations of the traditional DEA model, the super efficiency model (SBM) is gradually becoming a popular research topic [48]. The model mainly focuses on the following problems. When multiple efficiency values are 1 (DEA effective), the traditional model cannot further distinguish the efficiency difference of 1 unit, but SBM allows the efficiency value to be greater than 1, and scientific research can further accurately distinguish the unit efficiency value [49]. Combined with the defect of the traditional model being unable to calculate the undesired output, this paper intends to combine the unexpected output model with the SBM model to form a model that can overcome two problems at the same time, as follows:

$$\min \mathsf{\rho }=\frac{1+\frac{1}{\mathrm{m}}{{\displaystyle \sum }}_{\mathrm{i}=1}^{\mathrm{m}}{\mathrm{A}}_{\mathrm{i}}^{-}/{\mathrm{x}}_{\mathrm{ik}}}{1-\frac{1}{{\mathrm{q}}_1+{\mathrm{q}}_2}\left({{\displaystyle \sum }}_{\mathrm{r}=1}^{{\mathrm{q}}_1}{\mathrm{A}}_{\mathrm{r}}^{+}/{\mathrm{y}}_{\mathrm{rk}}+{{\displaystyle \sum }}_{\mathrm{t}=1}^{{\mathrm{q}}_2}{\mathrm{A}}_{\mathrm{r}}^{\mathrm{b}-}/{\mathrm{b}}_{\mathrm{rk}}\right)}$$

(1)

$$\mathrm{S}.\mathrm{t}.{\displaystyle \sum }_{\mathrm{j}=1,\mathrm{j}\ne \mathrm{k}}^n{\mathrm{x}}_{\mathrm{ij}}{\mathsf{\lambda }}_{\mathrm{j}}-{\mathrm{A}}_{\mathrm{i}}^{-}\le {\mathrm{x}}_{\mathrm{ik}}$$

(2)

$${\displaystyle \sum }_{\mathrm{j}=1,\mathrm{j}\ne \mathrm{k}}^{\mathrm{n}}{\mathrm{y}}_{\mathrm{ij}}{\mathsf{\lambda }}_{\mathrm{j}}-{\mathrm{A}}_{\mathrm{r}}^{+}\ge {\mathrm{y}}_{\mathrm{rk}}$$

(3)

$${\displaystyle \sum }_{\mathrm{j}=1,\mathrm{j}\ne \mathrm{k}}^{\mathrm{n}}{\mathrm{b}}_{\mathrm{tj}}{\mathsf{\lambda }}_{\mathrm{j}}-{\mathrm{A}}_{\mathrm{t}}^{b-}\le {\mathrm{b}}_{\mathrm{rk}}$$

(4)

$$1-\frac{1}{{\mathrm{q}}_1+{\mathrm{q}}_2}\left({\displaystyle \sum }_{\mathrm{r}=1}^{{\mathrm{q}}_1}{\mathrm{A}}_{\mathrm{r}}^{+}/{\mathrm{y}}_{\mathrm{rk}}+{\displaystyle \sum }_{\mathrm{t}=1}^{{\mathrm{q}}_2}{\mathrm{A}}_{\mathrm{r}}^{\mathrm{b}-}/{\mathrm{b}}_{\mathrm{rk}}\right)>0$$

(5)

$${\mathsf{\lambda }}_{\mathrm{j}},{\mathrm{A}}_{\mathrm{i}}^{-},{\mathrm{A}}_{\mathrm{r}}^{+}\ge 0$$

(6)

$$\mathrm{i}=1,2,\cdots \mathrm{m};\mathrm{r}=1,2,\cdots \mathrm{q};\mathrm{j}=1,2,\cdots n;\left(\mathrm{j}\ne \mathrm{k}\right)$$

(7)

In Model (6), it is assumed that there are n decision-making units, and each decision -making unit has an input vector, an expected output vector, and an undesired output vector. Assuming that there are m types of inputs and $\mathrm{q}$ types of outputs, including ${\mathrm{q}}_1$ expected outputs and ${\mathrm{q}}_2$ undesired outputs, the input vector is $\mathrm{x}\in {\mathrm{R}}^{\mathrm{m}}$, the expected output vector is $\mathrm{y}\in {\mathrm{R}}^{\mathrm{q}1}$, and the undesired output vector is $\mathrm{b}\in {\mathrm{R}}^{\mathrm{q}2}$. Where S represents the slack of the input and output, ${\mathrm{A}}^{-}$ represents the input redundancy, ${\mathrm{A}}^{+}$ represents the expected output shortage, ${\mathrm{A}}^{\mathrm{b}-}$ represents the undesired output excess, $\mathsf{\lambda }$ is the weight vector, and $\mathsf{\rho }$ represents the efficiency score.

3.3. Variable Description

The dependent variable is green investment efficiency. Based on the SBM super efficiency DEA model, this paper takes green investment in steel production as the input, the discharge of waste (waste gas, wastewater, and waste residue) produced in steel production as the undesired output, and the output value of the three wastes’ utilization as the expected output. The calculated efficiency value is shown in

Table 3. Summary statistics.

Variables	Variable Symbol	Obs	Mean	Std. Dev	Min	Max
Green investment efficiency	Gie	5400	0.6216	0.8038	0.0293	1.6182
Pubic environmental attention	lnPea	5400	3.9305	3.6325	2.0049	6.0855
Environmental regulation	lnEr1	5400	8.7652	0.6725	8.2774	9.2974
Environmental responsibility	lnEr2	5400	2.0753	1.8525	0	3.4340
Business environment	lnBe	5400	0.9242	0.5724	0.0490	2.2160
Enterprise scale	lnScale	5400	15.012	0.692	11.642	17.302
Main business income	lnMbi	5400	14.9230	1.6016	3.2189	12.5022
Economic density	lnEd	5400	7.3123	1.2506	3.2581	10.7920
Regional carbon emissions	lnRce	5400	3.3142	3.4271	2.1689	4.1292

. The higher the index, the higher the green investment efficiency of the enterprise.

The independent variable is public environmental attention. According to previous studies, this paper uses the monthly average of the Baidu haze search index to describe the public’s environmental attention, the main reason for this being that: (1) as the largest Chinese search engine, Baidu has a wide coverage and high data availability, and its data can best reflect the actual situation; (2) Compared with other environmental topic keywords such as “environmental pollution”, “haze weather” has stronger environmental perception, as the public can identify smog weather only by visibility; (3) as PM2.5 pollution concentration is highly correlated with air quality, haze pollution is a good measure.

The influence mechanism variables are environmental regulation intensity and enterprise environmental responsibility: (1) environmental regulation intensity: this paper uses the frequency of environmental words in the government work report to measure the intensity of regional environmental regulation; (2) enterprise environmental responsibility: the measurement of enterprise environmental responsibility in this paper adopts the score of environmental responsibility in corporate social responsibility by Hexun (www.hexun.com accessed on 25 March 2022).

The moderating variable is the business environment. This paper uses the number of kilometers of roads in the region divided by the area of the region to measure the regional business environment (in prefecture-level cities).

The control variables include the following: (1) enterprise scale—the annual total output value (logarithm) of the enterprise is used to measure the scale of the enterprise; (2) main business income—it is measured by the total annual main operating income (logarithm); (3) economic density—the economic density is measured by dividing the total annual GDP in the region by the regional area (in provinces); (4) regional carbon emissions—elect the total annual carbon emissions in the region to measure the regional carbon emissions (in prefecture-level cities). The larger the index, the more regional carbon emissions and the worse the environment.

The original data were from the China Iron and Steel Association, “China Urban Statistical Yearbook”, “China Statistical Yearbook” (from recent years),the statistical yearbooks of various provinces, and Hexun (www.hexun.com accessed on 25 March 2022).

3.4. Model Building

In order to make an empirical analysis of the above hypothesis, this paper constructs relevant econometric models with reference to the practices of Cai et al. [50] to test the relationship between pubic environmental attention and enterprise green investment efficiency. Since the impact of public environmental attention is lagging, this paper chooses the enterprise green investment lagging behind for two periods to conduct empirical research. The models are as follows:

$${\mathrm{Cie}}_{\mathrm{it}-2}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{lnPea}}_{\mathrm{it}}+{\mathsf{\alpha }}_2{\mathrm{X}}_{\mathrm{it}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(8)

where i is enterprise and t is time, t − 2 represents two lag periods, Gie is the efficiency of monthly enterprise green investment, lnPea is the monthly degree of environmental public attention, X is the control variables, ${\mathsf{\alpha }}_{\mathrm{i}}$ is parameter to be estimated, and ${\mathsf{\epsilon }}_{\mathrm{it}}$ is the random disturbance term.

This paper adds business environment variables as well as constructs the interaction term between pubic environmental attention and business environment to examine the moderating relationship between pubic environmental attention and enterprise green investment efficiency by business environment:

$${\mathrm{Cie}}_{\mathrm{it}-2}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{lnPea}}_{\mathrm{it}}+{\mathsf{\alpha }}_2{\mathrm{lnBe}}_{\mathrm{it}}+{\mathsf{\alpha }}_3{\mathrm{lnPea}}_{\mathrm{it}}\ast {\mathrm{lnBe}}_{\mathrm{it}}+{\mathsf{\alpha }}_4{\mathrm{X}}_{\mathrm{it}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(9)

where lnBe is business environment, and other variables are the same. In order to explore the mechanisms of public environmental attention to enterprise green investment efficiency, this article constructs the mediating effect model [51]:

$${\mathrm{Me}}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{lnPea}}_{\mathrm{it}}+{\mathsf{\beta }}_2{\mathrm{X}}_{\mathrm{it}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(10)

$${\mathrm{Cie}}_{\mathrm{it}-2}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{lnPea}}_{\mathrm{it}}+{\mathsf{\gamma }}_2{\mathrm{Me}}_{\mathrm{it}}+{\mathsf{\gamma }}_3{\mathrm{X}}_{\mathrm{it}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(11)

where Me represents the influence mechanism. If ${\mathsf{\alpha }}_1$, ${\mathsf{\beta }}_1$, and ${\mathsf{\gamma }}_2$ are all significant, and ${\mathsf{\gamma }}_1$ is smaller than ${\mathsf{\alpha }}_1$ or becomes insignificant, this indicates that Me in this paper has a mediating effect.

4. Empirical Analysis

4.1. Benchmark Test

Combined with model (8) and model (10), this paper uses the fixed effect model to empirically test the relationship between public environmental attention and green investment efficiency, and further investigate the moderating impact of the business environment on the two variables.

Since the Baidu search index is divided into mobile search and personal computer (PC) search, this paper examines the relationship between the two variables from three aspects, namely the comprehensive index (the combination of mobile and PC), mobile index, and PC index.

According to

Table 4. Benchmark test.

Gie-2	Fixed Effect
	Composite Index		Mobile Index		PC Index
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.2394 **	0.3094 ***	0.3428 ***	0.3316 ***	0.2381 *	0.2612 **
	(2.52)	(2.80)	(3.36)	(3.08)	(1.89)	(2.28)
LnBe		0.1168 ***		0.1721 ***		0.1268 *
		(3.36)		(3.00)		(1.86)
LnPea × LnBe		0.0725 **		0.0624 **		0.0806 *
		(2.61)		(2.36)		(1.72)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	1.3398 **	4.2987 **	10.7521 ***	8.6352 ***	6.4240 ***	4.7429 **
	(2.28)	(2.22)	(4.00)	(4.26)	(3.90)	(2.50)
Hausman test	16.32	15.36	18.44	17.36	20.31	18.36
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE	FE	FE
${\mathrm{R}}^2$	0.2661	0.2912	0.3412	0.3192	0.2883	0.2643

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

, the three indicators all show that the coefficient of public environmental attention is significantly positive, indicating that public environmental attention plays a significant role in promoting enterprise green investment efficiency. The coefficients of the business environment are significantly positive, indicating that the business environment promoted the efficiency of the green investment of enterprises to a certain extent. The interaction coefficients between public environmental concern and business environment are significantly positive, indicating that the influence of public environmental attention on enterprise green investment efficiency is positively affected by the business environment, which verifies Hypothesis 1 and Hypothesis 2. In addition, the significant level of mobile index is significantly higher than that of the PC index.

4.2. Robust Test

In order to further verify the robustness, this article modifies the model in many aspects, such as the replacing method, changing the lag period, and eliminating the influence of factors. The regression results are as follows:

4.2.1. Replace Method

The maximum likelihood estimation is used to test the robustness.

Table 5. Robust test: replace method.

Gie-2	Mle
	Composite Index		Mobile Index		PC Index
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.3521 **	0.3137 ***	0.3852 ***	0.2982 ***	0.2204 **	0.2065 **
	(2.31)	(3.40)	(3.06)	(3.64)	(2.40)	(2.18)
LnBe		0.0916 **		0.1025 ***		0.0762 **
		(2.56)		(2.89)		(2.12)
LnPea × LnBe		0.0634 **		0.0756 ***		0.0386 *
		(2.04)		(3.27)		(1.78)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.9655 ***	1.7532 **	3.2613 ***	3.8323 **	2.2561 ***	2.8642 **
	(3.26)	(2.36)	(3.08)	(2.60)	(3.12)	(2.42)
LR chi	265.83	300.74	312.27	296.86	305.5	298.56
${\mathrm{R}}^2$	0.3022	0.3208	0.3268	0.3138	0.2947	0.2865

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

indicates that the results of the robustness test are consistent with Table 4, which demonstrates the robustness of the model to a certain extent.

4.2.2. Adjust the Lag Period

Since the influence of public attention is lagging, this paper plans to change the lag period to further test the robustness of the model. The benchmark test is the effect of lag period 2. In the robustness test, this paper plans to select lag period 1 and lag period 3 data.

Table 6. Robust test: Adjust lag period 1.

Gie-1	Fixed Effect
	Composite Index		Mobile Index		PC Index
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.2692 *	0.2586 **	0.3011 ***	0.2585 ***	0.1864 *	0.1942 **
	(1.90)	(2.01)	(2.98)	(3.06)	(1.72)	(1.98)
LnBe		0.1723 **		0.1965 ***		0.1367 **
		(2.26)		(3.15)		(1.98)
LnPea × LnBe		0.0883 **		0.1052 **		0.0 437 *
		(2.02)		(3.21)		(1.87)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.9273 ***	2.3742 ***	4.0725 ***	5.9442 ***	2.7556 ***	4.3268 **
	(3.52)	(2.94)	(3.21)	(2.87)	(3.08)	(2.42)
Hausman test	25.65	24.90	21.65	19.98	20,84	21.78
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE	FE	FE
${\mathrm{R}}^2$	0.3091	0.2882	0.2631	0.3012	0.2996	0.2827

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

and

Table 7. Robust test: adjust lag period 2.

Gie-3	Fixed Effect
	Composite Index		Mobile Index		PC Index
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.3826 **	0.2952 ***	0.3328 ***	0.3126 ***	0.2462 **	0.2631 **
	(2.56)	(2.88)	(3.02)	(3.15)	(2.37)	(2.08)
LnBe		0.2166 **		0.1893 ***		0.2910 *
		(2.52)		(3.80)		(1.76)
LnPea × LnBe		0.2814 **		0.2692 ***		0.3094 *
		(2.41)		(3.098)		(1.90)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	2.2176 ***	1.8386 ***	6.1298 ***	4.3762 ***	0.9176 ***	1.0824 **
	(3.14)	(2.90)	(3.88)	(4.02)	(3.66)	(2.27)
Hausman test	20.37	18.86	19.54	19.75	20.15	21.82
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE	FE	FE
${\mathrm{R}}^2$	0.3021	0.2852	0.3117	0.3096	0.2830	0.2761

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

show the data test of lag period 1 and lag period 3, respectively, the results of both of which are consistent with the benchmark test—further illustrating the robustness of the model.

4.2.3. Eliminate the Influence of Factors

Since some factors have potential influence and interference on enterprise green investment, this paper intends to eliminate these influences in various ways. The specific measures are as follows: Firstly, since the time span of the sample selected in this paper is 2009–2017, which is a period of the rapid development of the Internet and smart phones. there may be inaccurate attention to the public environment due to the low Internet access rate and low popularity of smart phones in some regions. Therefore, this paper excludes certain remote and disconnected areas (including all cities under the jurisdiction of Tibet, Inner Mongolia, and Xinjiang provinces) to eliminate this impact; Secondly, since excessive environmental regulation may affect enterprise green investment, according to the practice of Li et al. [5], this paper excludes certain enterprise samples engaging in excessive environmental regulation to eliminate the impact of excessive policies and regulations on green investment and further test the influence of public environmental attention on enterprise green investment efficiency; Finally, considering the synchrony of air pollutants, carbon market pilot policies may also affect air quality [52]. This paper deletes the sample of enterprises in carbon pilot cities, and further tests the influence of public environmental attention (composite index) on enterprise green investment efficiency.

Table 8. Robust test: eliminate the influence of factors.

Gie-2	Fixed Effect
	Drop Remote and Backward Areas		Drop Strong Environmental Regulation City		Drop Carbon Market Pilots
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.3845 ***	0.3527 ***	0.2856 **	0.2254 **	0.3027 ***	0.3285 ***
	(3.14)	(3.30)	(2.28)	(2.31)	(2.98)	(3.16)
LnBe		0.2842 ***		0.2954 ***		0.2673 **
		(3.60)		(3.06)		(2.58)
LnPea × LnBe		0.1284 **		0.1092 **		0.0862 **
		(2.40)		(2.21)		(2.38)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	3.7654 *	2.7524 **	1.9148 ***	3.0063 ***	2.3219 ***	2.9754 **
	(1.76)	(2.01)	(3.20)	(3.72)	(3.37)	(2.48)
Hausman test	20.72	21.64	19.86	19.82	20.36	18.84
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)
Model selection	FE	FE	FE	FE	FE	FE
${\mathrm{R}}^2$	0.2792	0.3031	0.2980	0.3134	0.2871	0.3075

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

indicates the result of the three situations of eliminating the influence of factors that are consistent with the benchmark test, which further illustrates the robustness of the model.

The results in Table 5, Table 6, Table 7 and Table 8 show that the coefficients of public environmental attention are significantly positive. The coefficients of the business environment are significantly positive. The interaction coefficients between public environmental attention and business environment are significantly positive. The results of all robustness tests are basically consistent with the benchmark test results, which also verifies that the model setting is robust to some extent.

4.3. Dynamic Panel Test

In order to further solve the endogenous problems existing in the model setting, that is, the causal relationship between independent variables and dependent variables, as well as the potential sample selectivity bias and missing variable bias, this paper introduces the three-order lag term of green investment efficiency as a tool variable to build a dynamic panel model, and uses the system GMM method to correct some errors in the static panel model.

The results in

Table 9. Dynamic panel test: system GMM.

Gie-2	System GMM
	Composite Index		Mobile Index		PC Index
	Model I	Model II	Model I	Model II	Model I	Model II
Gie-3	0.3255 ***	0.3533 ***	0.2895 ***	0.3029 ***	0.2845 ***	0.2648 ***
	(3.68)	(3.36)	(4.00)	(3.88)	(3.12)	(3.31)
LnPea	0.2071 **	0.1984 **	0.428 **	0.2652 **	0.1989 *	0.2092 *
	(2.60)	(2.35)	(2.50)	(2.54)	(1.82)	(1.70)
LnBe		0.2012 *		0.2831 *		0.2165
		(1.80)		(1.88)		(1.56)
LnPea × LnBe		0.0724		0.0538		0.0622
		(1.28)		(1.39)		(1.46)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.9012 *	0.8374	1.3638	0.9837	1.2877 *	1.0936
	(1.78)	(0.98)	(0.71)	(1.39)	(1.89)	(1.06)
AR (1)	0.0782	0.1023	0.0518	0.0612	0.0787	0.0925
AR (2)	0.0589	0.0862	0.0712	0.0539	0.0619	0.0712
Sargan	27.9837	22.4731	19.7812	21.8731	26.8631	18.8361
p	0.0671	0.0876	0.1034	0.0871	0.0646	0.0682

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

show that all AR(2) values are greater than 0.05, and the model does not have the problem of second-order autocorrelation; the p-values are all greater than 0.05, and the model does not have the problem of transition identification. The results of the dynamic panel are consistent with Table 4 (except for some variables that are not significant).

The results in Table 9 show that the lag coefficient of enterprise green investment is significantly positive, indicating that the lag period of enterprise green investment has a significant positive correlation with the current period. The coefficients of public environmental attention are significantly positive. The coefficients of business environment are significantly positive. The interaction coefficients between public environmental attention and business environment are significantly positive, indicating that public environmental attention and business environment can promote the efficiency of green investment, and the influence of public environmental attention on enterprise green investment efficiency is positively affected by business environment. All the test results are basically consistent with the benchmark test, which not only verifies the robustness of the model, but also to a certain extent solves the endogeneity problem of the model settings.

4.4. Mechanism Analysis

The previous analysis indicates that public environmental attention can promote enterprise green investment efficiency through two transmission mechanisms, namely environmental regulation and environmental responsibility. Based on the benchmark regression, this paper further adds two mediating variables to verify the previous analysis and hypothesis.

The results in

Table 10. Mechanism analysis.

	Fixed Effect
	lnEr1	Gie-2	lnEr2	Gie-2
LnPea	0.0963 ***	0.0276	0.1283 ***	0.0671
	(3.28)	(0.90)		(1.36)
lnEr1		0.1372 ***
		(2.82)
lnEr2				0.2191 ***
				(3.02)
Control variables	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes
Constant	5.9273 ***	4.9274 ***	2.9375 ***	2.9743 **
	(3.05)	(4.01)	(3.82)	(2.21)
Hausman test	16.81	18.43	20.01	21.09
	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE
${\mathrm{R}}^2$	0.3226	0.2937	0.2591	0.2930

Note: **, and *** indicate significance at the 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

show that the coefficients of public environmental attention and mediating variables in the two regressions are significantly positive, the coefficient sign of the model empirical results is consistent with the expected theoretical sign in the model construction, and the mediating variable test is passed. This shows that public environmental attention promotes the efficiency of the green investment of enterprises through environmental regulation and environmental responsibility as mediating variables, which also verifies H3 and H4 to some extent.

4.5. Heterogeneity Analysis

Since the characteristics of different types of samples are inconsistent based on different environments, this paper intends to conduct a classification test based on three aspects—scale, ownership, and region—to investigate the heterogeneity of public environmental attention. The results are as follows.

4.5.1. Scale Heterogeneity Test

Since both financial decision-making and green environmental protection practices are inevitably affected by the scale of the enterprise, the sample enterprises is divided into two groups according to the size of the output value of the enterprise to further test the influence of public environmental attention on the enterprise green investment efficiency of different scales.

Table 11. Scale heterogeneity.

Gie-2	Large Enterprises		Medium-Sized Enterprises
	Model I	Model II	Model I	Model II
LnPea	0.3592 ***	0.3018 ***	0.1839 *	0.2032 *
	(3.01)	(2.92)	(1.90)	(1.81)
LnBe		0.3041 **		0.2186 **
		(2.12)		(2.30)
LnPea × LnBe		0.0781		0.0826 **
		(1.50)		(2.27)
Control variables	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes
Constant	1.8643 ***	1.3572 ***	2.9527 ***	2.3913 **
	(3.85)	(3.23)	(3.08)	(2.40)
Hausman test	20.51	19.28	21.90	20.26
	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE
${\mathrm{R}}^2$	0.2892	0.2661	0.2782	0.2903

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

shows that, firstly, public environmental attention plays a more significant role in promoting the green investment efficiency of large enterprises than that of medium-sized enterprises (the statistical level of 1% for large enterprises and 10% for medium-sized enterprises is significant); secondly, the business environment has a great effect on promoting green investment efficiency; finally, the moderating effect of the business environment is only significant in medium-sized enterprises, but not in large-scale enterprises.

4.5.2. Ownership Heterogeneity

Due to the different purposes of enterprises with different ownerships, differences in ownership may affect enterprises’ decision-making and development. This paper divides the sample of enterprises into two categories: state-owned enterprises and private enterprises, and examines the heterogeneity of public environmental attention under different ownerships.

The results in

Table 12. Ownership heterogeneity.

Gie-2	State-Owned Enterprise		Private Enterprise
	Model I	Model II	Model I	Model II
LnPea	0.3207 ***	0.3126 ***	0.2003 *	0.2917 *
	(3.86)	(2.99)	(1.71)	(1.83)
LnBe		0.2842 **		0.1839 ***
		(2.16)		(3.34)
LnPea × LnBe		0.0824		0.0581 **
		(1.36)		(2.45)
Control variables	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes
Constant	3.4821 ***	5.2841 ***	3.8252 ***	6.2311 **
	(4.22)	(5.91)	(3.71)	(6.45)
Hausman test	19.27	25.81	21.24	20.38
	(0.0000)	(0.0000)	(0.0001)	(0.0000)
Model selection	FE	FE	FE	FE
${\mathrm{R}}^2$	0.3017	0.2839	0.2408	0.2691

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

show that, firstly, compared with private enterprises, public environmental attention plays a more significant role in promoting the green investment efficiency of state-owned enterprises; secondly, the business environment promotes the efficiency of green investment; finally, the moderating effect of the business environment only affects private enterprises, and has no significant impact on state-owned enterprises.

4.5.3. Regional Heterogeneity

The environmental protection requirements of different regions are inconsistent, which affects the green investment of enterprises. This paper intends to divide the sample enterprises into three groups according to their location—the eastern, central, and western groups—to test the heterogeneity of public environmental attention in different regions.

Table 13. Regional heterogeneity.

Gie-2	East		Center		West
	Model I	Model II	Model I	Model II	Model I	Model II
LnPea	0.3630 ***	0.2754 ***	0.1482 *	0.1864 **	0.3012 *	0.2843 *
	(2.88)	(3.02)	(1.70)	(2.00)	(1.89)	(1.82)
LnBe		0.2351 **		0.2165 **		0.2429 **
		(2.31)		(2.42)		(2.14)
LnPea × LnBe		0.0729		0.0812		0.0893 **
		(1.31)		(1.42)		(2.46)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes	Yes
Control region	Yes	Yes	Yes	Yes	Yes	Yes
Constant	1.4552 ***	2.2163 **	3.3281 ***	3.9121 ***	2.8164 **	1.7412 ***
	(2.90)	(1.98)	(2.77)	(3.32)	(2.26)	(3.02)
Hausman test	16.86	16.19	18.93	19.08	17.71	18.00
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)
Model selection	FE	FE	FE	FE	FE	FE
${\mathrm{R}}^2$	0.3063	0.3168	0.2884	0.3069	0.2649	0.2822

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively, and the numbers in brackets are “t” of the estimated coefficients.

shows that, firstly, the promotion of public environmental attention on green investment efficiency is more significant in eastern China than it is in central and western China; secondly, the promotion of business environment on green investment efficiency is significant in all regions; finally, the moderating effect of the business environment is only significant western China, not in the eastern and central China.

5. Discussion

In recent years, public and government have paid more attention to environmental protection. A good ecological environment is not only conducive to people’s healthy life, but also promotes the development of a harmonious society to a certain extent. As the source of pollution, enterprises must control pollution emissions while promoting economic growth.

Green investment has become the first choice for enterprise development. Through green investment, enterprises can not only improve the environment, but also improve their reputation. It is a positive reflection of corporate social responsibility and is very beneficial to the long-term development of enterprises [44]. Therefore, this paper selects Chinese large- and medium-sized steel enterprises to explore the relationship between public environmental attention and enterprise green investment efficiency, and analyzes the relevant research results as follows.

Firstly, this article tests the positive influence of public environmental attention on green investment efficiency. Strong public concern for the environment can exert some pressure on governments [5], force the government take action, and improve the environmental protection practices of enterprises [22]. This is also consistent with the conclusion of this study. With the increasingly serious pollution problem, it is not only the environmental problem itself, but also a series of social problems caused by environmental problems that urgently need to be solved. As a source of pollution, enterprises should take primary responsibility for pollution. When the public pays attention to the environment to a certain extent, enterprises (especially heavily polluting enterprises) not only produce a large number of pollutants, but also reduce their social reputation, especially in advocating green environmental protection, which will damage the enterprise brand to a certain extent. At this time, only by improving environmental responsibility, controlling its pollution emissions through green investment, and standing out among similar enterprises can enterprises obtain a better brand reputation [43,44]. The increasing attention being paid to the ecological environment shows that the public has high demands in terms of the ecological environment. At this time, the government can only exert pressure on enterprises through some operations such as strengthening environmental regulation, increasing environmental supervision, incorporating environmental assessment targets into the assessment system of government officials, forcing enterprises to increase their green environmental protection practices, and improve green investment efficiency. As such, it seeks to fundamentally reduce a series of problems caused by environmental pollution.

Secondly, this study examines the positive impact of the business environment on green investment efficiency. The impact of public environmental attention on green investment efficiency is positively regulated by the business environment. A good business environment provides a good production environment and sales platform for enterprises. Convenient transportation and preferential policies have always been the basic guarantee for enterprises to maximize profits [29]. A good business environment can make it easier for enterprises to make profits. Some studies have shown that a good business environment can significantly promote the development of enterprises [28]. Facing pressure the from public and the government, enterprises must make green investment if they want to change the past high emission production mode. However, green investment requires a large amount of capital investment and has no income in the short term [44]. Therefore, it is difficult for enterprises with low returns or enterprises with difficult financial conditions to implement green investment [40]. A good business environment can significantly improve the performance and financial situation of enterprises. When public pays great attention to the environment, a good business environment can improve the performance of enterprises so that enterprises have better green investment ability and achieve the effect of green investment efficiency.

Thirdly, this study mainly discusses two transmission mechanisms of public environmental attention affecting enterprise green investment efficiency, namely environmental regulation and environmental responsibility. Environmental regulation and environmental responsibility play a mediating role. When the public is highly concerned about environmental issues, both the government and enterprises are facing the pressure brought by public environmental attention. On the one hand, for the government, in order to relieve the public pressure, the government has taken measures such as strengthening environmental supervision to regulate the behavior of enterprises, forcing enterprises to improve the green investment efficiency [15]. When the public’s awareness of environmental protection is gradually improved, the enterprise begins to pay attention to its own green reputation, strives to improve its environmental responsibility, and controls the emission of pollutants by improving the efficiency of green investment within the scope of financial resources, so as to achieve the effect of improving environmental quality [44].

Finally, the study examines the influence of public environmental attention on the green investment efficiency of different types of enterprises. Based on the heterogeneity of scale, the influence of public attention on promoting the green investment efficiency of large enterprises is more significant than that of medium-sized enterprises (large enterprises have a significant statistical level of 1%, while medium-sized enterprises have a significant statistical level of 10%). However, the moderating effect of a business environment is only significant in medium-sized enterprises, but not in large enterprises. The main reason is that, on the one hand, green investment requires a lot of capital and technological investment, and large enterprises have better green investment capabilities by virtue of their capital and technological advantages. [53]. When the public pays more attention to environmental issues, in order to maintain their reputation, large enterprises can significantly increase green investment and promote green investment efficiency, while small enterprises do not have this ability; on the other hand, when the business environment is poor, large enterprises still have the ability to make green investments. However, the small enterprises can only just meet government management requirements, and thus its green investment efficiency is very low. Some small enterprises are even more willing to pay small fines instead of green investment [41]. When the business environment improves, the performance of small enterprises’ performance improves, and they have certain green investment capabilities. They can implement green investment, and the efficiency will be significantly improved. At this time, large enterprises have always insisted on green investment. Changes in the environment will not affect its green investment efficiency, resulting in an insignificant moderating effect of the business environment. Based on the heterogeneity of ownership, public environmental attention plays a more significant role in promoting the green investment efficiency of state-owned enterprises than private enterprises. This is because state-owned enterprises also play a role of macro-control in addition to production and operation. The production decisions of state-owned enterprises often need to be considered from the overall level of the state and society, sometimes even without considering the cost [54]. When the public is highly concerned about the environment and the government is under pressure, state-owned enterprises will take the lead in increasing investment in environmental protection and significantly improve green investment efficiency. The purpose of private enterprise is profit. When the business environment is relatively poor, and even public attention to environmental issues has brought pressure to enterprises, private enterprises are often less enthusiastic about low-income activities such as green investment than state-owned enterprises in terms of cost. With the improvement in the business environment, the operating conditions and economic conditions of private enterprises have also improved, leading to changes in their concepts. They will pay more attention to their own green reputation, so as to strengthen green investment and improve the environment. At this time, the moderating effect of the business environment is more significant in private enterprises. Based on regional heterogeneity, public environmental attention in the east has played a more significant role in promoting green investment efficiency. Due to the developed economy and numerous enterprises in eastern China, the pollution is relatively serious, and environmental supervision is very strict. Public environmental attention has placed greater pressure on government and enterprises. Almost all enterprises in the eastern region take their environmental performance very seriously. However, the environmental regulations of enterprises in other regions are relatively loose and pay less attention to environmental performance [55]. The effect of green investment is not as significant as that of enterprises in eastern China, and the pressure of public environmental concerns on enterprises is not as great as that of eastern enterprises, and its effect on promoting the green investment efficiency is not significant.

6. Conclusions and Recommendations

6.1. Conclusions

According to the panel data of steel enterprises in China, the article examined the relationship between public environmental attention and enterprise green investment efficiency, deeply discussed its transmission mechanisms, and further studied several heterogeneous characteristics of enterprises. The results are as follows.

Firstly, public environmental attention improves the overall green investment efficiency of steel enterprises; secondly, the business environment improves the green investment efficiency of steel enterprises, and the influence of public environmental attention on the green investment efficiency of steel enterprises is positively affected by the business environment; thirdly, public environmental attention affects the green investment efficiency through environmental regulation and environmental responsibility; finally, based on the heterogeneity of scale, public environmental attention has a more significant impact on the green investment efficiency of large enterprises than medium-sized enterprises. The moderating effect of the business environment is only significant for medium-sized enterprises, but not for large enterprises; based on the heterogeneity of ownership, public environmental attention has a more significant impact on the green investment efficiency of state-owned enterprises than private enterprises. The moderating effect of the business environment is only significant for private enterprises, not for state-owned enterprises; based on regional heterogeneity, the impact of public environmental attention on green investment efficiency in the eastern region is higher than that in the central and western regions. Based on the empirical results and discussions, this paper makes some policy recommendations.

6.2. Recommendations

The article examines the relationship between public environmental attention and the green investment efficiency of steel enterprises, and discusses the influence of public attention on green investment efficiency based on the heterogeneity of each enterprise. Based on the full text analysis, some policy recommendations are as follows.

Firstly, the government should further open up public information feedback channels. A large part of public attention is based on the fermentation of public opinion. If public attention to the environment can be effectively transmitted to government, it will help solve the relevant problems in a targeted manner, so as to do a good job in environmental governance.

Secondly, all sectors of society should enhance the public awareness of environmental protection. China has the largest manufacturing industry in the world, and pollution is everywhere. It is not enough to solely rely on the supervision of the relevant departments. Therefore, all sectors of society should strengthen environmental protection publicity, and increase the dissemination of health damage knowledge. The specific measures can be to strengthen community publicity and public service advertisements as well as compile environmental protection textbooks.

Thirdly, the government should standardize and improve environmental regulations and strengthen environmental supervision. When the government formulates environmental protection policies, it must take into account not only the health of the public environment, but also the affordability of enterprises, so as to realize the parallel between environmental protection and development. The government should further strengthen the environmental supervision of enterprises, advocate and encourage compliance with green development, and resolutely stop the violation of environmental laws and regulations.

Fourth, local governments should encourage enterprises to expand the scale of green investment and support enterprises’ green transformation. Green investment requires a lot of capital and human investment, and will not produce returns in the short term. Some enterprises, especially small and medium-sized enterprises, have difficulties in capital and technology. The government should give certain subsidies and preferential policies (such as tax relief, policy support, and environmental protection subsidies) to guide and promote the green transformation of enterprises.

Finally, enterprises should actively undertake the social responsibility of environmental protection. Enterprises should actively respond to environmental protection policies, increase the scale of green investment, and undertake environmental protection responsibilities when their own conditions permit.

Limitations and future research: (1) Due to the availability of data, this paper only examined the impact of public environmental attention on green investment in steel enterprises, and did not involve other industries; (2) For the measurement of public environmental attention, this paper only used the Baidu search index, and did not include the specific content of attention. Future research may involve other more specific issues of public environmental attention; (3) This study only involved Chinese enterprises, not foreign enterprises, especially in poor countries such as India. Whether public environmental attention is also effective is a topic worthy of discussion.