Can the Reform of the Transfer Tax System Affect Corporate Green Innovation—Evidence from China’s “BT to VAT” Reform

Abstract

Taxation has long been an important tool for dealing with externalities, and studying whether tax policies can be linked to today’s green development trend can help macroeconomic decisions influence the green transformation of enterprises. The article uses the business tax to VAT reform (BT to VAT), the largest turnover tax reform in recent years that took place in China from 2012 to 2016, as an example. The policy effect model was constructed using the difference in difference method to explore the driving effect of the tax reform and related variables on corporate green innovation. The results show that the policy effect of tax reform can effectively promote an enterprise’s green innovation behavior, with the promotion effect on invention-based green patents being significantly higher than that on utility-based green patents; there is a certain time lag in the promotion of green innovation by the policy effect of tax reform, with a significant effect starting from the third year of the policy. The green innovation behavior of enterprises is influenced by the characteristics of enterprises, with a significant positive correlation with enterprise size and operating capacity. Green innovation behavior is influenced by the characteristics of enterprises and has a significant positive correlation with the size and operating capacity of enterprises. The implementation of the policy of ‘reducing taxes and fees’ is conducive to green innovation, and an effective tax reform should be formulated gradually in response to the national conditions and market situation; on the basis of quality assurance, unnecessary steps in the process of green patent application and approval should be reduced to improve the efficiency of green innovation; effective social supervision or the appropriate realization of corporate equity can be adopted to promote green innovation. Corporate equity to promote corporate green innovation.

1. Introduction

In 2021, China’s ‘Outline of the 14th Five-Year Plan and 2035 Vision for National Economic and Social Development of the People’s Republic of China’ clearly pointed out that the green transformation of development methods should be accelerated and green development should be introduced into the national development process. In the face of the requirements of the new era of development, how to promote the deep integration of economic performance and green transformation has become an important issue to be solved in the sustainable and development of enterprises. Taxation is a policy tool that allows the state, in its capacity as a social manager, to impose a fixed and continuous remuneration on tax payers by virtue of its coercive power and has long been regarded as an important tool for dealing with externalities. As an important tool for macro-regulation, tax changes are bound to affect the behavioral choices of enterprises and are likely to have a profound impact on their green development. At present, domestic and international studies on tax reform and green development mainly focus on environmental protection taxes that target environmental regulation, such as energy taxes and carbon taxes, while there is a lack of studies on the impact of turnover taxes on corporate green development [1,2]. This paper attempts to explore the effect of tax reform on corporate green innovation from the perspective of transfer taxes.

From 2012 to 2016, China gradually implemented the largest turnover tax reform in recent years—the “BT to VAT” reform—from pilot to nationwide, changing taxable items previously subject to business tax to VAT, with the aim of reducing double taxation in the circulation of goods and improving the operational efficiency of enterprises. The “BT to VAT” reform is a major reform of the VAT system after the transformation of VAT and is an important component and measure of structural tax reduction. Initially, the “BT to VAT” reform was implemented in the transportation sector and then expanded to include the postal sector and some service industries; finally, in 2016, it was allocated to the construction, real estate, and financial sectors, achieving full coverage of Chinese industries. The “BT to VAT” reform not only reduces the tax base of enterprises in the transportation industry and eliminates double taxation, but also opens up the VAT deduction chain between manufacturing, the service industry, and the transportation industry, as well as within the transportation industry, so that enterprises can obtain more input tax deductions. The research of many scholars also shows that the “VAT to VAT” has a significant innovation incentive effect, and the reduction in the tax burden means that the retained cash flow of enterprises increases, which helps to alleviate the financial constraints of R&D and innovation activities. The enterprises implemented by the VAT reform have great differences in chronological order and individual characteristics. The transportation industry is the first batch of pilot industries, and the VAT chain of the industry is also relatively clear. Based on this, this paper takes the transportation industry, which has been fully included in the reform pilot in 2013 as the research object, and conducts in-depth research on the comprehensive “VAT reform” The impact of the pilot on the green development of enterprises and the analysis of whether there are differences in this impact have important theoretical and practical significance to provide reference for the optimization of macro reform plans and the adjustment of policy implementation paths.

2. Literature Review and Theoretical Mechanism

2.1. Literature Review

Green development is a historical process, first introduced by the British economist D. Pierce in The Blue Book of the Green Economy, which points to an environmentally friendly form of economic development. Since then, the academic community has conducted extensive and profound research on green development, and a more comprehensive research system has been formed, with the research mainly focusing on two levels: regional green economic development and enterprise green development. At the level of regional green economic development, existing research has focused on the examination of the green development of cities and their pathways. Urban development is a multifaceted and integrated process, and building low-carbon cities is currently an important part of urban planning, helping to promote green urban development [3,4]. In urban development, green behavior can be seen as a ‘calling card’ for the city and can lead the way in urban transport [5]. Urban green development is generally measured using urban green total factor productivity. Based on this, Wang and Tao measured the impact of low-carbon city policies on green total factor productivity growth in Chinese cities based on Chinese prefecture-level cities using a difference-in-difference model and a systematic GMM estimation method and showed that low-carbon city initiatives significantly contributed to green total factor productivity growth in cities and that this impact showed typical heterogeneity across regions [6]. Shao and Xia also measured urban total factor productivity based on Chinese prefecture-level cities and explored the impact of urban sprawl on green total factor productivity in China, concluding that moderate sprawl promotes green total factor productivity while excessive sprawl inhibits it [7]. Zhao et al. studied urban green development from the perspective of foreign direct investment (FDI) and found that FDI significantly inhibited the improvement of green total factor productivity in cities [8]. Green innovation is an important part of modern corporate development and competition, and whether or not a company has green innovation practices determines whether or not it engages in sustainable development. The performance of green innovation positively affects competitive advantage and significantly contributes to the sustainability activities of firms [9,10]. In a further study, Dai et al. found that the value gains from green innovation were more significant in the growth and decline periods of the life cycle [11]. For the pathway study, Meo et al. argued that green finance is an effective means of reducing corporate CO2 emissions [12]. Yu et al. used panel data from Chinese listed companies and found that financing constraints can hinder the green development of firms to some extent [13]. Skordoulis et al. found that green entrepreneurship promotes green growth in companies, with corporate strategy playing a mediating role [14]. Elsewhere, environmental protection taxes such as carbon and energy taxes also have an impact on green development [15,16].

VAT has always been an important tax in the tax system, and there has been widespread academic interest in the economic and financial effects of VAT. In recent years, scholars such as Kala et al., Olaoye et al. and Kala et al. have studied the utility of VAT in the economic sphere [17,18,19]. As for the tax reform, the “BT to VAT” reform policy implemented in China is currently the larger and more far-reaching tax reform, which has been studied by numerous scholars. Yu and Qi used a double-difference approach to study China’s business tax to VAT policy and found that the productivity of firms affected by the policy has increased significantly [20]. Bai and Wu find that the “BT to VAT” reform policy is effective in improving the investment efficiency of Chinese firms, increasing liquidity, and reducing firms’ reliance on external financing [21]. Cao et al. found a significant increase in output efficiency and substantive innovation among firms after the “BT to VAT” [22]. The “BT to VAT” policy has also increased the income distribution of residents [23]. As a turnover tax reform, “BT to VAT” involves all aspects of the business and is likely to have an impact at the business environment level. However, this part of the literature has rarely been studied, and only a few scholars have focused on it. Gao et al. studied the impact of China’s “BT to VAT” reform policy on total factor carbon efficiency in cities and found that “BT to VAT” reform significantly improved total factor carbon efficiency at the city level [24]. Zhou et al. found that China’s “BT to VAT” reform helped reduce the energy intensity of firms and achieved energy savings [25].

The review of the literature reveals that the research of domestic and foreign scholars on the green development of enterprises mainly focuses on the influencing factors of green development, and although the “replacement of business tax with VAT” is the largest and most far-reaching tax reform in recent years, few scholars have linked the policy effects of the “BT to VAT” reform to the green development of enterprises. In order to fill the gap in this research area, this paper attempts to use the data of Chinese listed enterprises in the transportation industry in Shanghai and Shenzhen A-shares from 2008 to 2019 to study the effect of the “BT to VAT” reform policy and the green development of enterprises using the double difference method and analyse the impact of the implementation of the “BT to VAT” reform policy on the green development of enterprises, with a view to enriching the research in this area.

2.2. Theoretical Mechanism

VAT is a transfer tax based on the value added during the flow of goods or services. In terms of the taxation principle, VAT is levied on the value added in the production, distribution, and services of goods, or on the added value of goods. Compared with business tax, VAT is taxable at all levels and deductible at all levels, which can effectively reduce the tax burden on enterprises and enhance their profitability [26]. From a national perspective, the full implementation of “BT to VAT” is an important means of deepening the reform of the fiscal and taxation systems and promoting economic restructuring and industrial transformation. This paper argues that the implementation of the “BT to VAT” reform is likely to promote green innovation in enterprises. On the one hand, if the profitability increases for enterprises, they are likely to further promote their transformation and development. The green transformation of enterprises takes into account operational costs and requires sufficient economic support. If the green transformation cannot bring economic benefits to the enterprises or even harm them, they will not carry out the transformation. The implementation of the “BT to VAT” reform reduces the unnecessary taxes paid by enterprises, which boosts their economic performance and helps them to have the ability to make the green transformation [27]. On the other hand, in the context of the country’s vigorous promotion of supply-side structural reform and the promotion of green corporate development, corporate behaviour is under great public opinion pressure, and by carrying out tax reforms that help reduce taxes at this point in time, companies are likely to incorporate green innovation strategies into their corporate development strategies in order to respond to national policies and improve their own positive public opinion. It can be seen that the implementation of the “BT to VAT” reform will promote green innovation in enterprises, whether it is an active behavior gained from cost reduction or a non-active behavior driven by the general context. Based on the above analysis, this paper proposes the following hypothesis:

H1. 

The “BT to VAT” reform has promoted green innovation in business.

On the one hand, there are the direct impacts of industrial transfer on the well-being of residents. The green transformation of enterprises is still in its infancy, and the “BT to VAT” reform has promoted green innovation and provided creativity for green products at enterprises. The green patent system is a special support for green science and technology from the enterprise intellectual property system, which can be used to measure the achievements of enterprises in green development. Green patents can be divided into green invention patents and green utility patents according to their nature, both of which can measure creative and different outcomes. Green invention patents require outstanding substantive features and significant progress, while green utility patents only require ‘substantive features and progress’ and require a much lower level of inventiveness than inventions [28]. Theoretically, green invention patents have relatively more advantages in terms of difficulty in application and in promoting the green development of enterprises. At the early stage of green development, there are fewer basic achievements that can be copied, modified, or updated, while there are more new achievements that can significantly contribute to green development. Based on this, the paper proposes the hypothesis that, at the early stage of green development, there are fewer basic achievements that can be copied, modified, or updated, while there are more new achievements that can significantly contribute to green development. Based on this, the paper proposes the hypothesis that:

H2. 

“BT to VAT” reform has a greater impact on invention-based green patents than on utility-based green patents.

It is likely that there is a time lag in the promotion of green innovation behavior by enterprises. On the one hand, since the pilot implementation of the “BT to VAT” reform in Shanghai Transport Enterprises on January 1 2012, it has been gradually promoted by industry and region, and even in the transport industry, which was the first to complete full policy coverage, it took one year, which shows that the “BT to VAT” reform policy has a long process of achieving completeness. The slow advancement of the “BT to VAT” reform policy makes it difficult for enterprises to complete the adaptive reform of their internal financial systems within a short period of time, so the response time of development strategies to the “BT to VAT” reform policy may be prolonged, and the promotion effect of the “BT to VAT” reform on green innovation behavior may show a lagging characteristic. On the other hand, even for enterprises in the same industry, their differences are not to be underestimated due to the inconsistency of their main business, production environment, and business cycle, so the speed of response to the policy and the difficulty of the response procedure vary. Take the transportation industry as an example: to achieve green development, the transportation industry mainly requires environmentally friendly infrastructure and clean and low-carbon transport equipment, which is essentially a green transformation of fixed assets. However, the financial situation of an enterprise determines the way in which fixed assets are handled. Some enterprises with good financial status and sufficient capital are likely to scrap old fixed assets and renew machinery and equipment in advance in order to respond positively to national policies and enjoy policy dividends. Some of the enterprises with poor financial status, on the other hand, cannot afford to support the green transformation of their equipment before the depreciation and scrapping of fixed assets. Accordingly, this paper puts forward the research hypothesis that:

H3. 

There is a time lag in the promotion of green innovation behavior of firms by the “BT to VAT” reform.

3. Materials and Methods

3.1. Sample Selection and Data Sources

Given that the implementation of the “BT to VAT” reform policy in China has differences in chronological order and individual characteristics, this paper selects transport enterprises that were the first to fully implement the “BT to VAT” reform in 2013 as the treatment group and enterprises that implemented the “BT to VAT” reform in 2016 as the control group to test the impact of the “BT to VAT” reform policy on enterprises’ green innovation behavior. This paper selected Chinese listed A-share enterprises in Shanghai and Shenzhen from 2008 to 2019 as the initial research sample and excluded enterprises with discontinuous data during the study period and those with ST, ST*, suspended, and terminated listing during the study period to finally obtain 53 enterprises in the transportation industry and 44 enterprises as the control group, which were obtained from the CMRDS database, in order to ensure the stability of the experiment. Data for the remaining variables was obtained from the China Research Data Service Platform (CNRDS) database, with individual data taken from company annual reports.

3.2. Variable Setting

The explained variables in this paper are the green innovation behaviour of firms ($Green$). Green innovation behaviour measures the extent to which companies attach importance to environmental protection in their development process. The importance of enterprises to green development is often reflected in green R&D investment, and an important indicator to measure the transformation efficiency of green R&D investment is green patents, so this paper selects green patents as the measurement indicators of the green development of enterprises. In our country, green patents refer to patents on products or designs that are beneficial to the environment, adding consideration for environmental protection to the patent system’s emphasis on the creation, promotion, and application of science and technology. The data on green patents are divided into the number of green patents applied for in the year, the number of green patents obtained in the year, and the number of green patents granted in the year. Since green patents require a certain period of time to be tested and annual fees are paid, this paper selects the number of green patents applied for ($Greapp$) and the number of green patents granted ($Gregra$) and the sum of the two ($Green$) as indicators of green innovation behaviour, taking into account the stability and correctness of the study.

Since the double difference method (DID) has been widely used in policy effect evaluation in recent years, this paper intends to use the DID to evaluate the green development ability of enterprises with the interpreted variables. The basic principle of the double difference method (DID) is to use double difference estimators to exclude the effect of “pre-experimental differences” in the policy-affected experimental group and the non-policy-affected control group and then evaluate the effect of policy on the experimental group. The pilot reform model of “replacing business tax with VAT” by region, industry, and step by step can be regarded as a natural experiment, in which the enterprises included in the pilot areas and industries at the same time are the experimental group, and the other enterprises are the control group.

The core explanatory variable in this paper is the “BT to VAT” reform policy effect ($D$). This paper regards the implementation of the “BT to VAT” reform policy as a quasi-natural experiment implemented on a national scale. As the DID method has been commonly used to assess policy effects in recent years [29,30], this paper proposes to use the difference in difference method to assess the explained variable of corporate green innovation behavior. In this paper, grouping dummy variables $treat$ and time dummy variables $time$ are set according to the implementation characteristics of the “BT to VAT” reform policy. The experimental subjects were 53 listed companies in the transportation industry, which served as the treatment group for the experiment, assigned a value of $treat=1$. The 44 listed companies that implemented the “BT to VAT” reform in 2016 served as the control group for the experiment, assigned a value of $treat=0$. According to the implementation steps of the “BT to VAT” reform, it was implemented in 2012 in the transportation industry enterprises in Shanghai, Beijing, Jiangsu, and Zhejiang provinces in China, and the pilot “BT to VAT” reform in the transportation industry was extended to the whole country on August 1, 2013. The time dummy is assigned to $time=0$ for the period 2008–2013, and $time=1$ for the period 2014–2019, and the policy effect ($D$) is the interaction of the group dummy ($treat$) and the time dummy ($time$), i.e., $D=treat\ast time$.

Considering that the sample of the study is listed as companies in China, the indicators affecting the size and profitability development of the company are selected as control variables in this paper, as follows:

(1) Business maturity ($Age$). It is a measure of the length of time a company has been in existence. The higher the maturity of a company, the longer it has been in the market. This paper uses the time from the date of incorporation to the balance sheet date each year, divided by 365 days per year, as a measure of the maturity of a business.

(2) Enterprise size ($lnS$). Enterprise size refers to the degree of concentration of productivity and means of production in an enterprise and can be used as one of the considerations in specifying strategic planning for an enterprise, which generally adjusts in response to each size of enterprise reached. This paper uses the logarithm of the total assets of the enterprise at the beginning of each year as a measure of the size of the enterprise.

(3) Equity concentration ($ShrHolder3$). Shareholding represents the combined power of the company’s shareholders over the company’s people and property. Through shareholding, shareholders are qualified to manage the company and direct the management of the business. Considering the voice and control of the company’s management, this paper uses the shareholding ratio of the top 3 shareholders of the company as an indicator of equity concentration.

(4) Gearing ratio ($Aslbrt$). The gearing ratio is an important indicator for evaluating the solvency of an enterprise, reflecting how much of the enterprise’s total assets in its business activities are raised through debt and measuring the level of debt and the degree of risk of the enterprise. The higher the gearing ratio, the greater the degree to which the company is operating with debt. When the gearing ratio reaches or exceeds 100%, the company as well as having no net assets or being insolvent.

(5) Return on Assets ($ROA$). ROA is an important indicator to evaluate the profitability of an enterprise, reflecting the level of profit harvested from all assets and measuring the profitability of the enterprise’s use of assets. Generally speaking, the higher the ROA of an enterprise, the better the level of input and output of the enterprise, and the more effective the operation of the enterprise’s assets.

(6) Truthfulness in reporting ($4Big$). Accounting firms are important players in the social monitoring of enterprises, performing “due diligence” by issuing audit reports on the economic situation of enterprises to investors, creditors, and society as a whole to evaluate the business situation of enterprises. In order to evaluate the authenticity of the audit report, this indicator is measured by whether the company is audited by one of the Big Four international accounting firms (PricewaterhouseCoopers, Deloitte, KPMG, and Ernst & Young).

(7) Ownership ($Equity$). The ownership of Chinese enterprises can be divided into state-owned enterprises and non-state-owned enterprises, with state-owned enterprises being those that are funded by the state and whose entire property belongs to the state. Generally speaking, these enterprises enjoy a higher industrial status and economic level and are more likely to respond to national policies, so this indicator is measured by whether the enterprise is state-owned. The variables involved in this paper are shown in

Table 1. Variable definition.

Variable Type	Variable Name	Variable Symbols	Variable Definition
Explained variables	green innovation behaviour	$Greapp$	Number of green patent applications in the year (1)
		$Gregra$	Number of green patents obtained that year (2)
		$Green$	(1) + (2)
Explanatory variables	the “BT to VAT” reform policy effect	$D$	$treat\times time$
Controlling variables	Business maturity	$Age$	Number of days the company has been in existence ÷ 365
	Enterprise size	$LnS$	Total assets at the beginning of the year taken as a logarithm
	Equity concentration	$ShrHolder3$	Shareholdings of top three shareholders
	Gearing ratio	$Aslbrt$	(Total liabilities ÷ Total assets) × 100%
	Return on Assets	$ROA$	(Net profit ÷ average total assets) × 100%
	Truthfulness in reporting	$4Big$	1 for the Big Four audit; 0 otherwise
	Ownership	$Equity$	1 for state ownership; 0 otherwise

.

3.3. Model Setting

Based on the above variables, the fixed effects models (1)–(3) are constructed in this paper as follows.

$$Greapp=D+Age+Lns+ShrHolder3+Aslbrt+ROA+4Big+Equity+\epsilon$$

(1)

$$Gregra=D+Age+Lns+ShrHolder3+Aslbrt+ROA+4Big+Equity+\epsilon$$

(2)

$$Green=D+Age+Lns+ShrHolder3+Aslbrt+ROA+4Big+Equity+\epsilon$$

(3)

where $\epsilon$ is a constant term, the coefficient of $D$ in Equations (1)–(3) is expected to be positive, indicating that the implementation of the “BT to VAT” reform policy helps to promote green innovation in enterprises.

4. Empirical Results and Analysis

4.1. Descriptive Statistics

Table 2. Variable descriptive statistics.

Variable	N	Mean	p50	SD	Min	Max	Range
$Green$	636	2.321	0	7.333	0	66	66
$Greapp$	636	1.458	0	4.673	0	43	43
$Greach$	636	0.863	0	2.878	0	27	27
$age$	636	18.11	17.59	9.689	3.833	120.1	116.2
$LnS$	636	23.12	23.01	1.372	18.49	26.23	7.743
$ShrHolder3$	636	0.231	0.230	0.116	0.00100	0.533	0.531
$Aslbrt$	636	0.476	0.437	0.324	0.0580	5.778	5.720
$Roa$	636	3.957	4.025	12.37	−216.1	127.1	343.2
$4Big$	636	0.209	0	0.407	0	1	1
$Equity$	636	0.943	1	0.231	0	1	1

provides descriptive statistics for the variables covered in this paper. A total of 636 observations were made for 53 enterprises in the transport sector from 2008–2019. The explained variable of corporate green innovation behavior is that, on average, the number of corporate green patent applications in the year is 1.458 times greater than the number of corporate green patents obtained in the year 0.863, indicating that some patents do not meet the green patent conditions when applying for green patents. The mean values of the control variables enterprise maturity, enterprise size, equity concentration, gearing ratio, and return on assets are all very close to the median. The distribution of these variables is basically symmetrical, with no left or right bias, indicating that there are very few extreme values in the study sample and that the sample is relatively robust.

4.2. Analysis of Regression Results

Table 3. Regression results on the policy effects of “BT to VAT” reform and green innovation.

Variables	(1)	(2)	(3)
	$\mathit{G}\mathit{r}\mathit{e}\mathit{a}\mathit{p}\mathit{p}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{g}\mathit{r}\mathit{a}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{e}\mathit{n}$
$D$	0.721 ***	1.166 ***	1.886 ***
	(3.34)	(3.27)	(3.48)
$Age$	−0.006	0.025	0.020
	(−0.46)	(1.25)	(0.64)
$lnS$	0.276 *	0.541 **	0.817 **
	(1.73)	(2.06)	(2.04)
$ShrHolder3$	−4.251 ***	−9.207 ***	−13.457 ***
	(−3.96)	(−5.20)	(−4.99)
$Aslbrt$	1.048 ***	1.629 ***	2.677 ***
	(3.03)	(2.86)	(3.08)
$Roa$	0.010	0.015	0.025
	(1.44)	(1.30)	(1.43)
$4Big$	−1.127 ***	−2.289 ***	−3.416 ***
	(−3.10)	(−3.81)	(−3.74)
$Equity$	−0.015	−0.137	−0.151
	(−0.03)	(−0.19)	(−0.14)
Constant	−4.871	−9.623	−14.495
	(−1.32)	(−1.58)	(−1.56)
Observations	1164	1164	1164
R-squared	0.417	0.428	0.446
Company FE	YES	YES	YES
Year FE	YES	YES	YES

Note: t statistic in brackets; *, ** and *** respectively represent significant at the significance level of 10%, 5% and 1%. The following tables are the same.

shows the regression results of the “BT to VAT” reform policy effect and corporate green innovation. The results show that the explanatory variables of the “BT to VAT” reform policy effect ($D$) have a positive effect on the explained variables of green patent application ($Greapp$), green patent acquisition ($Gregra$) and the sum of the two ($Green$) at the 1% level of significance, indicating that the implementation of VAT reform policy can significantly improve the green innovation of enterprises, so the hypothesis H1 is valid. Due to the stimulus of the “BT to VAT” reform policy, the input tax of enterprises can be deducted, so the investment in R&D fixed assets will increase, which will be reflected in the results. Furthermore, the promotion effect of the “BT to VAT” reform and enhancement on the acquisition of green patents is 1.166, which is higher than its promotion effect on the application for green patents of 0.721. Comparing with the descriptive statistics, the number of green patent applications by enterprises is greater than the number of green patents obtained by enterprises, which indicates that “BT to VAT” reform and enhancement helps enterprises’ invention patents meet the requirements of the green patent system more, thus promoting the acquisition of green patents. In addition, the control variables enterprise size and asset-liability ratio both have a significant positive effect on enterprises’ green innovation behavior, especially when the gearing ratio is significant at the 1% level, indicating that enterprises’ solvency is positively correlated with their willingness to engage in green innovation, and the stronger their solvency, the more likely they are to engage in green innovation, which further illustrates that enterprises’ green innovation behavior is based on their business conditions. However, the control variables equity concentration and truthfulness in reporting both significantly reduce firms’ green innovation at the 1% level. From a management perspective, the goal of a company is to maximize shareholder wealth, and promoting green development in a company is likely to reduce the company’s profits and affect the interests of shareholders, which shareholders do not want to see, so the indicator of equity concentration presents a hindering effect. The higher the authenticity of the financial report, the more transparent the company is. The reason why the authenticity of financial reports significantly reduces the green development of companies may be that it is more difficult for companies to falsify, alter, or embellish corporate indicators, resulting in the loss of some of the ‘grey’ benefits and therefore reducing green innovation behavior.

According to H2, this paper collects data on the number of corporate green invention patent applications ($Greappin$), corporate green utility patent applications ($Greappus$), corporate green invention patent applications ($Gregrain$), and corporate green utility patent applications ($Gregraus$), respectively, and regresses them with the effect of the “BT to VAT” policy;

Table 4. Regression results on “BT to VAT” reform and different nature of green patents.

Variables	(4)	(5)	(6)	(7)
	$\mathit{G}\mathit{r}\mathit{e}\mathit{a}\mathit{c}\mathit{h}\mathit{i}\mathit{n}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{a}\mathit{c}\mathit{h}\mathit{u}\mathit{s}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{g}\mathit{r}\mathit{a}\mathit{i}\mathit{n}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{g}\mathit{r}\mathit{a}\mathit{u}\mathit{s}$
$D$	0.033	0.142	0.263 **	0.196
	(1.18)	(0.96)	(2.21)	(1.06)
$Age$	−0.001	−0.008	−0.006	−0.011
	(−0.53)	(−0.95)	(−0.84)	(−1.03)
$lnS$	−0.001	−0.184 *	−0.010	−0.218
	(−0.03)	(−1.68)	(−0.12)	(−1.58)
$ShrHolder3$	−0.142	−2.127 ***	−1.374 **	−2.302 **
	(−1.02)	(−2.89)	(−2.32)	(−2.49)
$Aslbrt$	0.032	0.339	0.110	0.374
	(0.72)	(1.43)	(0.57)	(1.26)
$Roa$	0.001	0.006	0.002	0.006
	(0.65)	(1.23)	(0.46)	(1.00)
$4Big$	−0.143 ***	−1.818 ***	−0.961 ***	−2.233 ***
	(−3.03)	(−7.29)	(−4.78)	(−7.13)
$Equity$	−0.048	−0.388	−0.330	−0.347
	(−0.86)	(−1.33)	(−1.40)	(−0.94)
Constant	0.162	5.863 **	1.397	6.826 **
	(0.34)	(2.32)	(0.69)	(2.15)
Observations	1164	1164	1164	1164
R-squared	0.269	0.334	0.331	0.321
Company FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES

Note: t statistic in brackets; *, ** and *** respectively represent significant at the significance level of 10%, 5% and 1%. The following tables are the same.

shows the regression results of the VAT policy effect and the different nature of green patents. It was found that when regressed separately, the “BT to VAT” reform policy effect only had a positive effect on the number of green invention patent applications ($Greappin)$ by enterprises at the 5% significant level, indicating that the implementation of the “BT to VAT” reform promoted green invention patent applications by enterprises to a certain extent, in line with hypothesis H2. However, the reason why only one of the four regressions is significant deserves deeper investigation. This paper argues that, firstly, from the perspectives of application (6) and acquisition (4), green invention patents are inventive, and although the application process is procedurally easier, because they are exploratory green achievements, they will face a more stringent and complicated examination process, which leads to a significant promotion of green invention patent applications but not for the acquisition of green invention patents. From the perspective of the nature of green patents (6) and (7), green utility patents are simpler in their research process as they are adaptations and updates of existing achievements, and the implementation of the VAT reform policy is likely to have only a minor impact on their applications, whereas green invention patents are creations and require more investment due to the complexity of the research process, so they are more significantly affected by the “BT to VAT” reform. Furthermore, under the rules of green patent examination, method-based inventions cannot be used to apply for green utility patents, and inventors are more likely to file applications for green invention patents when there is ambiguity in identifying the category of research results, which is the reason for the significant boost in the number of green invention patent applications due to the “BT to VAT” reform.

4.3. Parallel Trend Test

When using the DID method to assess the policy effect of a policy, it is necessary to test whether there is no significant difference between the treatment and control groups before the point in time when the policy occurs. If the trend is similar or consistent between the treatment and control groups before the point in time when the policy occurs, then the sample passes the parallel trend test, and the double difference method can be used to assess the policy effect. Figure 1, Figure 2 and Figure 3 show the parallel trend tests for the explanatory variables green patent applications ($Greapp$), green patent acquisitions ($Greach$) and the sum of the two ($Green$), respectively.

From Figure 1, Figure 2 and Figure 3, it can be seen that before the implementation of the “BT to VAT” reform policy, neither the treatment group nor the control group showed a large difference. Furthermore, after the implementation of the policy, the difference between the treatment and control groups widened as the time point increased, indicating that the empirical evidence in this paper passed the parallel trend test and was stable. Furthermore, Figure 1, Figure 2 and Figure 3 consistently show a gradual widening of the difference between the treatment and control groups after the policy implementation point in time, starting from the third period after the policy implementation (post_3), indicating that the significant effect of the “BT to VAT” reform policy effect has a time lag rather than an immediate effect from the period of occurrence, confirming hypothesis H3.

4.4. Placebo Test

In order to enhance the causal effect of this paper, a placebo test is conducted using the ‘construct pseudo-policy’ method. Referring to the existing literature, this paper uses the panel data of the original treatment group and the control group from 2008 to 2012, assuming that the ‘pseudo-policy’ occurred in 2010. The group dummy ($Ftreat$) and time dummy ($Ftime$) of the difference in difference model were reset, assigning the treatment group dummy $Ftreat=1$ and the control group dummy $Ftreat=0$; the time dummy variable $Ftime=0$ before 2010, otherwise $Ftime=1$ and the interaction term $FD=Ftreat\ast Ftime$. Regression results are shown in

Table 5. Placebo test.

Variables	(8)	(9)	(10)
	$\mathit{G}\mathit{r}\mathit{e}\mathit{g}\mathit{r}\mathit{a}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{a}\mathit{p}\mathit{p}$	$\mathit{G}\mathit{r}\mathit{e}\mathit{e}\mathit{n}$
$FD$	0.019	0.288	0.307
	(0.13)	(1.23)	(0.95)
$lnS$	0.007	−0.126	−0.119
	(0.05)	(−0.50)	(−0.35)
$ShrHolder3$	−2.525 **	−3.785 *	−6.310 **
	(−2.07)	(−1.87)	(−2.27)
$Aslbrt$	0.271	0.335	0.606
	(0.63)	(0.47)	(0.61)
$Roa$	0.003	0.003	0.006
	(0.50)	(0.32)	(0.45)
$4Big$	−0.827 **	−1.986 ***	−2.813 ***
	(−2.57)	(−3.71)	(−3.83)
$Equity$	−0.026	−0.105	−0.131
	(−0.07)	(−0.17)	(−0.15)
Constant	0.945	4.695	5.639
	(0.28)	(0.85)	(0.74)
Observations	485	485	485
R-squared	0.487	0.441	0.527
Company FE	YES	YES	YES

Note: t statistic in brackets; *, ** and *** respectively represent significant at the significance level of 10%, 5% and 1%. The following tables are the same.

.

A fixed-effects regression was conducted by moving the policy to 2010. The results show that when the ‘pseudo-policy’ occurred in 2010, there was no significant effect of the explanatory variable policy effect on the explained variables green patent application ($Greapp$), green patent acquisition ($Gregra$) and the sum of the two ($Green$), which can indicate that this paper has some causal effect in the regression of the policy effect of “BT to VAT” reform and enterprise green development. The regressions on the effect of the “BT to VAT” reform policy and the green innovation behavior of enterprises have some causal effect. In contrast, the inhibitory effect of the control variables equity concentration (ShrHolder3) and truthfulness in reporting (4Big) on green innovation is persistent and still shows a significant negative effect in the regression results, further illustrating the causal logic of this paper’s empirical evidence.

5. Conclusions and Recommendations

This paper examines the impact of the “BT to VAT” reform policy on corporate green innovation by using data from 2008–2019 on Chinese Shanghai and Shenzhen A-share listed transportation enterprises and related enterprises, using green patents as the explained variable, constructing the “BT to VAT” reform policy effect through the difference in difference method, conducting an empirical study on the implementation of the “BT to VAT” reform policy and corporate green innovation behavior, and examining the nature classification of green innovation behavior. The empirical results show that: (1) the implementation of the “BT to VAT” reform policy significantly promotes corporate green innovation, and the size and profitability of enterprises have a significant positive effect on green innovation, while the concentration of equity and the truthfulness of corporate financial reports have a significant inhibiting effect on green innovation. (2) The “BT to VAT” reform policy is more likely to promote the development of invention-based green patents because the research difficulty of invention-based green patents and utility-based green patents is different. (3) There is a certain time lag in the promotion of green innovation by the “BT to VAT” reform policy, and enterprises will be affected by the policy three years after its implementation. The findings of this paper show that the implementation of the “BT to VAT” reform policy helps enterprises carry out green innovation, thus facilitating the integration of economic performance and green transformation, which is of great significance in accelerating the realisation of green modernisation and promoting harmony between man and nature.

Based on the above findings, this paper tries to put forward the following three policy recommendations: (1) The taxation system is an important means to influence the behavior of enterprises. In order to promote the green transformation of enterprises macroscopically, we can start from the perspective of taxation and adopt the method of tax reduction to promote the green behavior of enterprises. When implementing the tax reform, it is necessary to take into account the business situations of enterprises and eliminate ‘one size fits all’. It is recommended that the taxation scheme be gradually improved to reduce the pressure on the business development of enterprises caused by the exclusion of rapid transformation. For countries that have not yet implemented VAT, the government can gradually or completely implement a VAT turnover tax system according to national conditions so as to achieve the effect of green transformation in enterprises by reducing the taxation of goods in circulation. (2) We will formulate supporting policies to encourage green technology innovation. A single turnover tax reform cannot guarantee the incentive effect on enterprises’ green technology innovation, which is superimposed with other environmental policies and innovation policies. In general, we will enhance the effectiveness of environmental laws, regulations, and standards. Secondly, we should actively use economic and incentive policies, formulate encouraging and voluntary policies, and strengthen the power of “internal restraint” in social and economic development.Finally, the formulation of green science and technology policies encourages chemical enterprises to develop new available resources, new materials utilization, clean production technology, environmental pollution treatment technology, and waste recycling and recycling technology, giving chemical enterprises their own technical advantages, making them the main force in pollution control, and guiding our economy to achieve high-quality development. (3) On the non-tax side, the professional requirements for accounting firms to be diligent and responsible to society in providing true and effective corporate financial reports should be improved, which will help enterprises develop in a transparent manner. Furthermore, by reducing the concentration of shareholding in the enterprise, it will prevent the enterprise from becoming a ‘monopoly’ and facilitate the enterprise to use part of its profits for green development.