Study on the Impact of Public Attention Relative to Green Development Policies on the Return on Investment of Related Industries ^†

Abstract

As an emerging development model, green development requires the participation of the government, the market, and the public. It needs the formation of a positive interaction between policy guidance, market demand, and public will in order to realize a win–win situation comprising sustainable economic development and environmental protection. Thus, we explored green-development-related policies in China and defined four areas: industrial energy preservation, petrochemicals, agriculture, and the integration of the Yangtze River Economic Belt. Then, we analyzed how public attention relative to green development policies impacted the return of stock indices in the green industry using the Baidu search index. The research study was conducted on all trading days of A-shares in China from 1 January 2013 to 1 January 2023. In total, 2430 valid datasets were observed and curated over these 10 years. The public attention to green development policies was classified into four categories: industrial energy efficiency, petrochemicals, agriculture, and the Yangtze River Economic Zone. The attention focused on a category was measured by the number of searches of keywords that were the category’s names on the Baidu platform. The number of searches was considered the variable representing the public’s attention relative to each area of green development policies. The results showed that the increase in public attention to green development policies was not correlated with the increase in the return on investment (ROI) of green and carbon-neutral industries. In addition, information on green development policies had no significant effect on encouraging individuals to participate in investments in the green industry and had little effect on the financing of the carbon-neutral industry. The lack of details for green development policies and effective guidelines for potential investors needs to be overcome in order to encourage investment and financing.

1. Introduction

With the continuous development of the green industry, investors have paid attention to investment opportunities in the industry. Accordingly, the impact of the public’s attention to green development policies on the return on investment (ROI) of the green industry has been explored extensively. Although many studies have shown a positive relationship between public attention and ROI, the increase in public attention did not necessarily have a positive effect on the ROI and even showed a negative impact. Therefore, it is necessary to elaborate on this topic systematically.

For the study, all government documents from 2008 to January 2023 related to “green development” were collected and summarized to understand the direction of policy development and sort the documents according to the frequency of each direction. The top three items with a frequency of more than 15 were “industrial energy conservation”, “oil and petrochemical”, and “Yangtze River Economic Belt”. The policies in these three directions were selected as the research content in this study, and the corresponding Baidu search index was used to gauge public attention relative to the policies. Based on the collected content of green development policies, the stock prices of related companies in Tonghuashun market conditions were compiled to judge whether the stocks corresponded to the scope defined in the “Catalogue” in this study. “Carbon neutral”, “oil and petrochemical”, and “Yangtze River Economic Belt” were used as keywords to obtain the necessary stock information on the green industry.

As the unit of ROI and the Baidu search index were not identical, it was necessary to standardize their factors in order to build the model. Therefore, the changes in the ROI and the three-day moving average of changes in the Baidu search index were investigated for the model. In the model, the changes in the closing price of the Shenzhen Composite Index and the dummy variable of seasonal factors were selected as control variables, while the change in the ROI was used as the dependent variable. Variables were used to establish an ordinary least squares (OLS) regression model, and the effect on the model was analyzed based on the p-value and R².

2. Theories and Research Hypothesis

2.1. Definition of Concepts

The earliest document on green development policies was released in 2008. The Chinese government announced the 12th Five-Year Development Plan in 2009 [1]. China’s first green development plan included a development plan that could strengthen energy conservation and emission reduction, develop a green economy and industries, build a green China, and promote green civilization. In the “green development” policy from 2008 to the present, the three aspects most frequently mentioned were industrial energy conservation, petrochemicals, and the Yangtze River Economic Belt. Subsequent research and reviews have been based on these three aspects, which were also used as the perspective of public attention in this study.

2.1.1. Public Attention

Public attention refers to the degree of public attention to a policy of a country. After a policy is announced, the response to this policy needs to be understood in order to propose timely adjustments accordingly. Thus, public attention needs to be straightforward and effective for such adjustments. It also affects the stock price of related industries. This study was conducted based on this concept to determine the impact of public attention relative to green development policies on the ROI of the green industry.

2.1.2. Industrial Energy Efficiency

The green development policy is related to industrial energy efficiency and has been developed over the years. Methods for evaluating energy efficiency have become an important indicator for analyzing the effect of China’s industrial energy efficiency policies. Therefore, we included industrial energy efficiency as an indicator in this study.

2.1.3. Petrochemical Industry

The petrochemical industry is a pillar of China’s economy and has contributed to its economic development significantly. However, the industry considerably harms the environment. Thus, the transformation of the petrochemical industry into a green industry is required. Since 2012, China has issued policies for the green development of the petrochemical industry, which have received attention from the public and have been researched. Recently, a multi-perspective investor attention index based on Google Trends was proposed, and it considers four perspectives, including the linkage of alternative energy prices, energy prices, macroeconomics, and geopolitical factors, to enhance the predictive ability of oil prices and their fluctuations [2]. It was claimed that investors’ attention to macroeconomics and geopolitics affected oil prices.

2.1.4. Yangtze River Economic Belt

The Yangtze River Economic Belt is important in China’s major development strategies. Thus, there are many reports on the environmental governance and protection of the belt’s environment, showing significant improvements in environmental protection. We included public attention relative to the “Yangtze River Economic Belt” to study the impact of environmental protection on the ROI.

2.2. Theoretical Background

2.2.1. Ordinary Least Squares (OLS) Regression

OLS regression is a linear least squares method that selects unknown parameters in linear regression models by minimizing the sum of squared differences between the observed dependent variables. The least squares method shows the optimized and substantial effect observed in the Gauss–Markov theorem. The OLS model has been widely used. In Ref. [3], the relationship between parents and children’s height was studied using the concept of “regression to the mean”, which was the first research study that adopted the OLS model. OLS regression was used in this study for constructing the model to explain the impact of public attention on the fluctuation of closing stock prices.

2.2.2. Capital Asset Pricing Model (CAPM)

CAPM is based on the theory that the expected return of an asset is related to the market risk factor and irrelevant to other factors. However, the CAPM model cannot explain the fluctuation of the ROI of assets and related factors. To solve these problems, multi-factor models were proposed. In multi-factor models, it is assumed that the return rate of an asset is related to market risk factors, company size, book–market ratio, liquidity, and others. Risk factors are used to explain the volatility and differences in asset returns. A three-factor model [4] and a four-factor model (also known as the Fama–French three-factor model) [5] were proposed earlier. Since then, more multi-factor models, such as five- and six-factor models, have been proposed. Stock prices are influenced by multiple factors. Thus, Fama [6] proposed the efficient market hypothesis in which stock prices were explained with all available information, and the trend of stock prices was determined to be random and unpredictable. Malkiel [7] criticized the efficient market hypothesis, claiming that factors including fundamentals, technical factors, and psychology also influenced the trend of stock prices. Finally, Shiller [8] proposed using excessive volatility in stock prices based on the observation that the trend of stock prices exceeded the range of changes. Therefore, based on the summary of previous study results, public attention was used to explore the impact of public attention on stock prices.

2.3. Hypothesis for the Influence of Public Attention Relative to Green Policies

As environmental issues are increasingly recognized as serious, the public is paying more attention to environmental protection, and enterprises are taking responsibility for environmental protection. Therefore, the public’s attention to green development may affect the stock price of the green industry, which needs more investigation. Improving energy efficiency can promote the green development of the industry. The government’s practice of sustainable development and the improvement of the petroleum industry promote green development, especially in the Yangtze River Economic Belt. Therefore, green development is closely related to the green industry. We have selected two distinct methods for collecting data for public attention: one from PC users and the other from mobile phone users. Therefore, each aforementioned hypothesis will be analyzed with respect to the mutual impact on the stock price of the green industry based on public attention data from both PCs and mobile platforms. Based on this, the following hypotheses were proposed in this study.

Hypothesis 1.

The public’s attention to “industrial energy efficiency” based on pc platforms and the ROI of the industry related to “industrial energy efficiency” are mutually influential.

Hypothesis 2.

The public’s attention to “industrial energy efficiency” based on mobile platforms and the ROI of the industry related to “industrial energy efficiency” are mutually influential.

Hypothesis 3.

The public’s attention to “the petroleum industry” based on pc platforms and the ROI of the industry are mutually influential.

Hypothesis 4.

The public’s attention to “the petroleum industry” based on mobile platforms and the ROI of the industry are mutually influential.

Hypothesis 5.

The public’s attention to the “Yangtze River Economic Belt” based on pc platforms and the ROI of “environmental protection” are mutually influential.

Hypothesis 6.

The public’s attention to the “Yangtze River Economic Belt” based on the mobile platforms and the ROI of “environmental protection” are mutually influential.

3. Research Design

3.1. Variable Selection

The three-day moving average of the Baidu search index was calculated using Equation (1).

$$Baidu search rate of change=\frac{\left\{\frac{\mathrm{BD}\_\mathrm{Index}\left(T-2\right)+\mathrm{BD}\_\mathrm{Index}\left(T-1\right)+\mathrm{BD}\_\mathrm{Index}}{3}\right\}-\mathrm{BD}\_\mathrm{Index}\left(T-2\right)}{\mathrm{BD}\_\mathrm{Index}(T-2)}$$

(1)

The three-day moving average (Baidu search change rate) was selected as a variable along with the change in the closing price and seasonal variables of the Shenzhen Composite Index. The change in the green industry stocks’ closing price was chosen as a dependent variable. Control variables included seasonal changes and the changes in the Shenzhen Composite Index closing price.

3.2. Sample Data

Data from the Baidu search index were collected from 22 March 2013 to 31 December 2022, except for closed days of the market. The three-day moving average was calculated to obtain the Baidu search change rate, and the one-day forward shift was calculated from the PRE_Baidu search change rate. The change in the ROI and the Shenzhen Composite Index’s closing price were obtained from Tonghuashun. The seasonal dummy variables were compiled as follows: spring (0) from March to March, summer (1) from April to June, autumn (2) from July to September, and winter (3) from October to December. The final sample time range is from 27 March 2013 to 30 December 2022.

3.3. OLS Regression Model Construction

The three sets of variables were summarized into the following three models.

$$TZH=\alpha +{\beta }_1\mathrm{pre}TZH+{\beta }_{2}Index\_pc+{\beta }_{3}preIndex\_pc+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(2)

$$TZH=\alpha +{\beta }_1\mathrm{pre}TZH+{\beta }_{2}Index\_yd+{\beta }_{3}preIndex\_yd+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(3)

$$SH=\alpha +{\beta }_1\mathrm{pre}SH+{\beta }_{2}Inde{x}_{pc}+{\beta }_{3}preInde{x}_{pc}+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(4)

$$SH=\alpha +{\beta }_1\mathrm{pre}SH+{\beta }_{2}Inde{x}_{yd}+{\beta }_{3}preInde{x}_{yd}+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(5)

$$CJ=\alpha +{\beta }_1\mathrm{preCJ}+{\beta }_{2}Index\_pc+{\beta }_{3}preIndex\_pc+{\beta }_{4}SZ+{\beta }_{5}preSZ+\beta 6 Season+\epsilon$$

(6)

$$CJ=\alpha +{\beta }_1\mathrm{preCJ}+{\beta }_{2}Index\_yd+{\beta }_{3}preIndex\_yd+{\beta }_{4}SZ+{\beta }_{5}preSZ+\beta 6 Season+\epsilon$$

(7)

TZH, SH, and CJ represent the stock board returns of “carbon neutrality”, “petroleum and petrochemical”, and “Yangtze River Economic Belt” industries, respectively. preTZH, preSH, and preCJ denote their respective previous day’s stock board returns. Index signifies the Baidu search variation rate, while pre-index refers to its previous day’s search variation rate. SZ stands for the Shenzhen Composite Index stock board returns, and preSZ represents the previous day’s Shenzhen Composite Index stock board returns. Season indicates the seasonal index.

4. Empirical Analysis

4.1. Descriptive Analysis of Variables

“Industrial Energy Conservation” and “Carbon Neutrality”

On 21 March 2013, the China Nonferrous Metals Industry Association announced industrial energy conservation and green development. Provincial and municipal governments responded actively and issued their industrial green development plans. Since 2013, the national government has issued an industrial green development plan each year, and data from the Baidu search index with the keyword “industrial energy conservation” are shown in Figure 1.

The number of searches for “industrial energy conservation” reached its maximum between May and September 2014. As relevant policies during that period, the State Council issued the Action Plan on 15 May 2014. This plan aimed to achieve annual reductions of 3.9% in energy consumption per unit of GDP, 2% in chemical oxygen demand, 2% in sulfur dioxide, 2% in ammonia nitrogen, and 5% in nitrogen oxide emissions set in 2014 and 2015. Additionally, the plan targeted a 4 and 3.5% reduction in carbon dioxide emissions per unit of GDP in the two years. Since its introduction, this plan garnered significant public attention and response. In addition, the official website of the Chinese government provided an in-depth analysis of related policies. Consequently, there was a substantial increase in the number of searches.

Considering the emphasis on industrial energy conservation, the industrial chain related to carbon neutrality was represented with the stock price of the green industry. Furthermore, the trends in the closing prices of the stocks showed a lag compared to changes in public attention. This suggested the influence of public attention on the trends of the ROI.

The petrochemical industry is important and has high industrial relevance and a wide range of product coverage in China. However, as the economy develops, the demand for ecological and environmental protection increases, necessitating the green transformation of the petrochemical industry. In 2012, governments at various levels recognized this issue and began taking measures for the improvement of urban green development. The Yangzhou Municipal Government issued “Opinions” on 10 September 2012, outlining their plans. Subsequently, in 2016, the Hubei Province released the Petrochemical Industry Transformation Plan. The number of related searches with the keyword “petrochemical” is presented in Figure 2. The number of searches related to “petrochemical” remained relatively stable over time. However, there were periods when the number of searches increased. As the petrochemical industry has always been a subject of great interest and received attention from the public and government, changes in policies or significant events related to the industry led to a sudden surge in public attention.

In the stock market, there were 47 companies listed in the category of the petrochemical industry. In the period around June 2015, stock prices fluctuated when public attention to “petrochemicals” increased and declined. This showed a correlation between stock prices and the level of public attention relative to the petrochemical industry.

Since September 2013, the “Yangtze River Economic Belt” gradually gained attention from the public. The Belt covers 11 provinces and municipalities, including Shanghai, Jiangsu, Zhejiang, Anhui, Jiangxi, Hubei, Hunan, Chongqing, Sichuan, Yunnan, and Guizhou, spanning the eastern, central, and western regions. They have significantly contributed to China’s economic growth. On 21 July 2013, during an inspection visit to Wuhan, General Secretary Xi Jinping emphasized the strengthening of cooperation in the Yangtze River basin for inland water transportation and transforming the entire basin into a “golden waterway”. Figure 3 displays the number of searches with the keyword “Yangtze River Economic Belt”.

In September 2014, the Chinese government designated the Yangtze River Economic Belt as a national strategy for the construction of the economic belt based on the “golden waterway”. Around this time, there was a peak in public attention and interest in the Yangtze River Economic Belt, as indicated by the number of searches, which reached an average of over 20,000 searches per day. This period had a significantly higher number of searches than other periods. Due to the focus on environmental protection and ecological conservation in the construction of the Yangtze River Economic Belt, the “environmental protection” industry was chosen to represent the corresponding industry. The trend of the stock price was influenced by a notable increase in public attention around October 2014. Several months later, the stock prices of companies for environmental protection also began to increase. Therefore, there was a correlation between the two, but further empirical verification was necessary to determine the specific impact.

4.2. Analysis of Regression Results

In this section, two distinct methods for collecting data for public attention were employed: one from PC users and the other from mobile phone users. This clear differentiation enhances the systematic presentation of our findings.

4.2.1. Industrial Energy Efficiency

The formula for the Baidu search index for “industrial energy efficiency” on the PC platform is as follows:

$$TZH=\alpha +{\beta }_1\mathrm{pre}TZH+{\beta }_{2}Index\_pc+{\beta }_{3}preIndex\_pc+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(8)

where Index_pc denotes the search variation rate for “industrial energy efficiency” on the PC platform.

The formula for the Baidu search index of “industrial energy efficiency” on the mobile platform is as follows:

$$TZH=\alpha +{\beta }_1\mathrm{pre}TZH+{\beta }_{2}Index\_yd+{\beta }_{3}preIndex\_yd+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(9)

where Index_yd represents the search variation rate for “industrial energy efficiency” on the mobile platform.

The Baidu search change rate was not statistically significant in the model, and there was no sufficient evidence to conclude that the rate impacted the ROI of the industry related to carbon neutrality. The OLS regression results for the Baidu search change rate for “Industrial Energy Efficiency” are presented in

Table 1. Result of the OLS model of the Baidu search change rate and ROI of companies related to industrial energy efficiency.

	Coefficient	T	p > |T|
PRE_TZH	0.0718	3.616	0.000
ROC	−1.413×10−5	−0.831	0.406
PRE_ROC	−2.948×10−5	−1.736	0.083
SZ	0.0105	83.786	0.000
SZ_PRE	0.0009	3.504	0.000
SEASON	−0.0003	−2.519	0.012
R-SQUARED	0.751
F-STAT	1195
PROB (F)	0.000

. The rate was not statistically significant, but the pre-search rate was significant at p = 0.1. This suggested public attention to “Industrial Energy Efficiency” and the ROI were correlated.

4.2.2. Petrochemical Industry

The formula for the Baidu search index of the “Petrochemical industry” on the PC platform is as follows.

$$SH=\alpha +{\beta }_1\mathrm{pre}SH+{\beta }_{2}Index\_pc+{\beta }_{3}preIndex\_pc+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(10)

The formula for the Baidu search index of “Petrochemical industry” on the mobile platform is as follows.

$$SH=\alpha +{\beta }_1\mathrm{pre}SH+{\beta }_{2}Index\_yd+{\beta }_{3}preIndex\_yd+{\beta }_{4}SZ+{\beta }_{5}preSZ+{\beta }_{6}Season+\epsilon$$

(11)

Based on the results of the OLS regression model (Equations (10) and (11)), the Baidu search change rate was not significant in the model, and no significant impact of the rate was observed on the ROI of the petrochemical industry. The OLS regression results for the Baidu ssearch change rate for “Petrochemical Industry” are presented in

Table 2. Result of the OLS modeling of the Baidu search change rate and ROI of companies related to petrochemical industries.

	Coefficient	T	p > |T|
PRE_TZH	0.0724	3.644	0.000
ROC	−1.906×10−5	−1.822	0.069
PRE_ROC	−6.11×10−6	−0.584	0.559
SZ	0.0105	83.766	0.000
SZ_PRE	0.0008	3.432	0.001
SEASON	−0.0003	−2.527	0.012
R-SQUARED	0.751
F-STAT	1194
PROB (F)	0.000

.

4.2.3. Yangtze River Economic Belt

The OLS model for the Yangtze River Economic Belt is as follows:

$$CJ=\alpha +{\beta }_1\mathrm{preCJ}+{\beta }_{2}Index\_pc+{\beta }_{3}preIndex\_pc+{\beta }_{4}SZ+{\beta }_{5}preSZ+\beta 6 Season+\epsilon$$

(12)

where Index_pc denotes the search variation rate for the “Yangtze River Economic Belt“ on the PC platform:

$$CJ=\alpha +{\beta }_1\mathrm{preCJ}+{\beta }_{2}Index\_yd+{\beta }_{3}preIndex\_yd+{\beta }_{4}SZ+{\beta }_{5}preSZ+\beta 6 Season+\epsilon$$

(13)

where Index_yd denotes the search variation rate for the “Yangtze River Economic Belt “ on the mobile platform.

The three-day Baidu search change rate was not significant, while the rate before the three days was significant at p = 0.1. A significant correlation was observed between the public’s attention to the Yangtze River Economic Belt and the ROI of the industries related to environmental protection. The OLS regression results for the Baidu ssearch change rate for “Petrochemical Industry” are presented in

Table 3. Result of the OLS modeling of the Baidu search change rate and ROI of industries related to the Yangtze River Belt and environmental protection.

	Coefficient	T	p > |T|
PRE_TZH	0.0750	3.771	0.000
ROC	−2.149 × 10−5	−0.560	0.576
PRE_ROC	−7.182 × 10−5	−1.870	0.062
SZ	0.0089	69.265	0.000
SZ_PRE	0.0009	4.286	0.000
SEASON	−0.0003	−3.469	0.001
R-SQUARED	0.676
F-STAT	823.7
PROB (F)	0.00

.

5. Discussions

In the constructed OLS regression models, the Baidu research change rate was not important for industrial energy efficiency. The overall explanatory power of the model was 70%, and the coefficient was negative, indicating that the rate influenced the ROI of the related industry. The rate affected the ROI of the petrochemical industry. The overall explanatory power of the model was 67.5%, suggesting that there was room for further improvement. The coefficients of the model were negative, indicating that public attention influenced the ROI. For the Yangtze River Belt and environmental protection, the model’s explanatory power was 67.5%, and the coefficients were negative, suggesting that public attention had a negative impact.

As control variables in the models, the Shenzhen Composite Index and seasonal dummy variables mitigated the influence on the ROI. These two variables were significant at p = 0.05 in all models, demonstrating that the relationship between independent and dependent variables in the model was not ruined by the two variables. The Shenzhen Composite Index is a stock price index compiled by the Shenzhen Stock Exchange, reflecting the price trends of all A- and B-shares listed on the exchange, and it is significantly influenced by other markets. As stock market trends are correlated with seasons, seasonal factors and the Shenzhen Composite Index helped the model estimate the relationship between independent and dependent variables more accurately.

6. Conclusions

We explored the relationship between the public’s attention to different green policies and the ROI of the green industry. The results showed that the higher the public’s attention to green policies, the higher the ROI. Conversely, an increase in the ROI stimulated public attention relative to green policies.

Improving industrial energy efficiency is important and necessary to achieve carbon neutrality. By adopting highly efficient energy technologies, equipment, and management measures, energy consumption in the industry can be reduced with lowered carbon emissions. Then, carbon neutrality can be achieved. Empirical models proved a mutual relationship between the increase in public attention to industrial energy efficiency and the ROI of the industry related to carbon neutrality. Therefore, the government needs to improve industrial energy efficiency and disclose more details about implementing green development. The results of this study suggested that as public attention increased, investors paid more attention to investment opportunities in the petroleum industry, which increased the ROI. Increased public attention prompts the government to introduce more policies for developing the petroleum industry and increase the ROI. The government must actively practice sustainable development plans, strengthen corporate responsibility, and improve the green development of the petroleum industry.

As a national development strategy, the development of the Yangtze River Economic Belt has significant implications for sustainable development. On the one hand, with the continuous increase in public attention to environmental issues, the investment demand for the environmental protection industry increases accordingly. Implementing environmental protection policies and measures in the Yangtze River Economic Belt impacts the performance of the related industry, thereby affecting its RIO. Therefore, the government must exploit stakeholder theory. In conclusion, the government must improve the green development system, actively promote environmental greening, and facilitate green development.

While green policies are readily available, the selection of corresponding industry stocks is challenging. Most policies were not directly related to the industry. Thus, industries that corresponded the most to the policies were chosen in this study. Thus, the results of the models may vary depending on the methods for choosing the industries. The industry sectors for the policies of industrial energy efficiency and the petroleum industry were relatively well matched, while the sector for the “Yangtze River Economic Belt” was not satisfactorily matched. Therefore, in-depth research is needed to develop more accurate models.