With the rise in emissions from agricultural, livestock, and industrial sectors (Elahi et al., 2024; Buntaine et al., 2024; Abbas et al., 2023; Yu et al., 2022; Kang and Silveira, 2021), the global climate continues to warm, and severe weather events such as floods and droughts are occurring frequently, severely impacting human survival and sustainable economic development. To protect their ecological environment and public health, some countries have adopted Green Technical Barriers to Trade (GTBTs) within international trade, these are trade measures, regulations, and standards, that restrict or prohibit the import of highly polluting products (Liu et al., 2020). However, as international trade gradually expands and governments become more involved in trade, GTBTs have become the most frequently used non-tariff barrier for creating trade friction in the international market (Liu et al., 2023). Unlike most non-tariff barriers, GTBTs have four main characteristics: rationality, crypticity, targeting, and effectiveness (Crowley et al., 2018). Therefore, some countries that promote trade protectionism in the name of environmental protection have set strict environmental standards to boycott imports of goods from other countries (Peng et al., 2024). According to the WTO’s annual Technical Barriers to Trade reports, there were 556 environment-related technical barriers to trade (TBT) notifications in 2021, accounting for approximately 1/4 of all TBT notifications. GTBTs’ role as sanction tools in international trade negatively impacts developing countries’ trade exports which cannot be ignored (Zhou et al., 2023).

As China continues its high-level opening-up to the outside world, green technological innovation (GTI) has become a new advantage to cultivate international economic cooperation and competition (Du et al., 2021). In recent years, China’s foreign trade has faced unpredictable uncertainties owing to trade protectionism, the coronavirus disease 2019 (COVID-19) pandemic, global climate action, and supply chain restructuring (Wible, 2021). Weakening international demand and increasing pressure to maintain stable growth in imports and exports have significantly weakened the positive impact on sustainable economic development (Yan et al., 2024). The impact of non-tariff barriers on China’s export trade has received widespread attention from academia. However, the existing literature mainly focuses on the effect of anti-dumping and countervailing on changes in export scale (Crowley et al., 2018), export product quality (Zhang, 2022), and supply chain efficiency (Grossman et al., 2024), among others, and insufficient attention has been paid regarding how GTBTs affect the GTI of export enterprises. Given China’s massive export volume and significant global market share, it is vulnerable to restrictions on GTBTs in international trade with Chinese enterprises lagging behind developed nations in green production and environmental protection practices (Chandra, 2016). Accordingly, GTBTs have become an important academic topic for researchers in the field of non-tariff barriers to trade and sustainable development.

Furthermore, to ensure the coordinated achievement of pollution prevention and high-quality economic development, the Chinese government proposed the low-carbon city pilot policy (LCCPP). Since it began in 2012, the scope of the pilot has continuously expanded; it now includes six provinces, 80 cities, and one region (Shi et al., 2024). The LCCPP represents a significant strategic measure by China to actively pursue green, sustainable, and innovative development. Low-carbon cities have strengthened clean production standards, energy-saving low-carbon technology standards, and carbon footprint evaluation standards, promoting the alignment of domestic environmental standards with international environmental standards (Emerick et al., 2016). Low-carbon cities also explicitly encourage enterprises to increase investment in green technology research and development (R&D), provide a favorable policy environment and innovative resources to support enterprises in conducting high-quality, high-tech, and high-value-added GTI, and promote the effect of pollution reduction and carbon reduction through green low-carbon technology R&D, thereby contributing to the achievement of the “dual carbon” goals (Zhou et al., 2022). Meanwhile, to avoid imposing carbon tariffs due to GTBTs from developing countries, export enterprises need to follow the global low-carbon development trend, stay close to the forefront of international green technology, and continuously achieve the GTI to adapt to the global low-carbon technology revolution and low-carbon industry transformation (Li et al., 2022). This ultimately ensures that the products or services produced by the enterprise meet the environmental requirements of international consumers and further promotes the sustainable development of export enterprises.

This study explores the impact of the LCCPP on the GTI of export enterprises with the effect of GTBTs. Scholars have studied the relationship between GTBTs and enterprise innovation but have not reached a consensus. First, GTBTs negatively impact enterprises’ innovation activities. Some studies have suggested that carbon tariffs increase trade costs and reduce enterprises’ expected earnings. Enterprises tend to reduce innovation investments to avoid the risks posed by trade risks (Chandra and Long, 2013). GTBTs also improve international market access conditions and limit opportunities for enterprises to access overseas markets and resources. Mayer et al. (2014) found that GTBTs cause market competition, price stimulation, and income inhibition effects, which negatively impact enterprise innovation. However, the theory of optimal allocation of resources and the resource-based theory suggest that GTBTs have positive effects on enterprise innovation. Pan et al. (2022) found that, in the face of trade barriers and external challenges, some enterprises with low production efficiency have retreated from international markets (Pan et al., 2022). Export enterprises that continue to operate in the original market increase their investment in R&D, which improves their total factor productivity and innovation capabilities to bear risks and occupy market vacancies. The GTBTs may also induce transnational capital flows and trade transfers (Liu et al., 2023). Export enterprises meet the demands of new markets by developing new products and improving market competitiveness, which reflect the process of enterprises seeking independent innovation and development to cope with GTBTs.

Little research has been conducted on the relationship between the LCCPP and GTI of export enterprise, and there are three common views in the academic community: promoting, inhibiting, and uncertainty theories. Some studies suggest that export enterprises continuously engage in technological and product innovation through scale and scope economy, competition, and learning effects to adapt to foreign environmental standards and consumer preferences while participating in international market competition. Chen et al. (2021) used the green patent data of listed companies and found that the constraints of the LCCPP improved the GTI willingness of export enterprises and promoted the GTI behavior of high-carbon industries through government support, public participation, and financing constraints (Chen et al., 2021). Du et al. (2021) studied the impact of GTI willingness from the perspective of environmental regulation, finding that strict environmental regulations stimulate enterprises’ enthusiasm for green innovation and that market-based environmental regulations more significantly impact green innovation willingness (Du et al., 2021). In addition, Chen et al. (2022) not only explored the role of the LCCPP in promoting green innovation in cities, but they also analyzed their spatial spillover effects. The results indicate that the LCCPP can accelerate the green innovation process in both local and surrounding cities and achieve better spatial spillover effects in high-level, large, and eastern cities (Chen et al., 2022). Different from the above conclusions of scholars, Tian et al. (2021) found that the LCCPP not only failed to positively influence the green innovation of export enterprises but also reduced the GTI level of the city where the enterprises were located (Tian et al., 2021). Peng et al. (2024) studied the “pollution haven hypothesis” and environmental regulation in China and pointed out that pollution-intensive industries can avoid environmental regulations by changing their location without improving their own green technologies (Peng et al., 2024). Qiu et al. (2021) have explored the effects of the command-and-control type of the LCCPP, and found that there have insufficient incentives to promote the development of green technologies (Qiu et al., 2021).

The previous studies have clear shortcomings. First, most studies regard the GTI as an intermediary variable to explore the impact on the export quantity, export quality, or export performance of enterprises through innovation channels, and do not consider the effect of the LCCPP on the GTI of export enterprises. Second, in the face of frequent trade friction and the development trend of global low-carbon trade, the innovation effect of GTBTs has not focused on export enterprises. Export enterprises are vulnerable to the effects of international environmental regulations and economic policy uncertainty. Based on existing research, this study focuses on China’s A-share listed export enterprises between 2007 and 2021, using the difference-in-differences (DID) approach to empirically explore the effects of GTBTs on China’s export enterprises’ GTI. Then, the LCCPP is taken as quasi-natural experiments to analyze whether the LCCPP can help export enterprises cope with the impact of the GTBT. The possible contributions of this study are as follows. First, based on trade cost theory, the framework incorporates enterprise GTI decisions, trade breadth, and government policy support, thereby expanding the micro-theoretical framework for how GTBTs and the LCCPP influence export enterprises’ GTI. Second, beyond the negative effects of GTBTs, this study explores government support for export enterprises’ GTI through the LCCPP and the moderating effect of lthe LCCPP with different policy instruments and intensities. This research provides a detailed demonstration for export enterprises to optimize innovation resource structures, increasing the high-tech, high-value-added green products to breaking through the GTBTs, thus promoting the GTI to the greatest extent with government support, increasing the breadth and depth of international market participation, and facilitating national low-carbon development and green transformation.

To maintain sustainable development and balance the relationship between trade and environmental protection, environmental protection clauses are increasingly being incorporated into agreements concerning the import and export of trade products, mandating strict environmental certification for imported products. Importing countries have established a range of technical standards and regulatory frameworks to limit or ban the import of foreign products, emphasizing the need to safeguard the ecological environment, natural resources, and human health. By increasing trade costs through tax collection, GTBTs have intensified financing constraints, which have a “destructive effect” on existing enterprises in the export market and a “fear effect” on new entrants (Chandra, 2016), this situation ultimately leads to a decline in the competitiveness of export products, a loss of competitive advantage in target markets, and may cause enterprises exit the international market or change the trade flow (Song et al., 2020).

Specifically, (1) GTBTs lead to loss of market access rights and even punishment for shutdown if export enterprises fail to meet the standards in the short term (Crowley et al., 2018). The existence of a “green technology threshold” suggests that export enterprises are unable to surpass GTBTs and choose to delayed export with lower productivity, a single product structure, and limited risk resilience. This behavior not only worsens the cash flow situation of enterprises, but also limits domestic companies’ access to cutting-edge GTI and participation in international technology cooperation. Consequently, it fails to grasp the update direction of GTI and the goal of export enterprises’ green technology R&D and innovation in time (Pan et al., 2022). (2) Export enterprises can circumvent GTBTs by trade diversion, enterprises do not need to modify or innovate their own technology. When enterprises encounter GTBTs set up by a certain country, in addition to actively adapting to maintain market access rights in that country, they can also change their export decisions, transfer exports to other countries without GTBTs, thereby diminishing their motivation to engage in GTI (Chen et al., 2021). (3) For enterprises aiming to meet cleaner production standards, the development of green production and pollution control technologies requires significant investment in research personnel and long-term financial support, and the advancement of GTI research and development cannot yield substantial returns in the short term. Furthermore, the purchase of new pollution control equipment, advanced green production facilities, and the import of green intermediates increase production costs, raising export prices to cover trade and green technology transformation costs leads to decreased product demand and reduced earnings (Shi et al., 2024). Consequently, this situation dampens the enthusiasm and initiative of enterprises to implement green technology R&D activities. (4) In response to negative signals from enterprises facing GTBTs, financial institutions are reluctant to bear trade risks and reduce credit availability, resulting in intensifying financing constraints (Liu and Ma, 2020). Under the pressure of cost fluctuations and uncertainty, increasing R&D investment to enhance a firm’s long-term innovative capabilities is considered a high-risk and low-return behaviors. Changes in cash flow and financing structure will lead export enterprises to increase cash flow holdings and reduce investment in green technology R&D to face the special risks.

Hypothesis 1

GTBTs negatively impact the GTI of Chinese export enterprises.

GTI activities are characterized by high risk, substantial investment, continuous processes, and uncertain outcomes. The government developed policy packages for low-carbon pilot cities to foster regional low-carbon development, including financial subsidies, talent policies, and tax relief. The GTI impacts of the LCCPP in coping with GTBTs for export enterprises are primarily reflected in the following: (1) From the perspective of R&D investment, under the guidance of the LCCPP, innovation resources disrupt the original distribution pattern, improve the supply of regional R&D resources, and accelerate knowledge and technology spillover effects (Chen and Wang, 2022). Enterprises obtain high-level human capital and R&D funds through the government to compensate for market failure and underinvestment in R&D activities and help enterprises realize the GTI by reducing R&D risks (Peng et al., 2021). (2) From the perspective of the learning-by-exporting effect, enterprises conduct a series of innovative activities, including understanding the environmental standards of developed countries, improving the added value of green products, introducing green intermediate products, and improving green production processes by contacting foreign competitive products, competitors, and suppliers (Hong et al., 2021). According to the demonstration, collaboration, and export learning effects, export enterprises acquire knowledge about green product characteristics, green technology, cleaner production processes, and cleaner management through information exchange, resource sharing, knowledge absorption, transnational cooperation, technology application, and achievement promotion, which is conducive to enterprises’ realization of the GTI (Du et al., 2021). (3) From the perspective of the market competitive effect and resource allocation among enterprises, the constraints of LCCPP may lead to the exit of high-energy-consumption and high-emission export enterprises from the market. At this point, low carbon export enterprises will likely increase investment in GTI through the “winner effect” and further occupy the capital, labor and technological resources released by the exiting enterprises. (4) From the perspective of alleviating financing constraints for enterprises, pilot cities will also implement green financial policies which effectively make up for the gap in green innovation funds for export enterprises (Liu and Ma, 2020). Enterprises receive support and convey to the outside world that they have more environmentally friendly products, which helps improve their business credit, allowing them to obtain required R&D funds from different channels and break through the GTBTs. Accordingly, the LCCPP help export enterprises realize the leap over of GTBTs through GTI (Figure 1).

Hypothesis 2

The LCCPP can alleviate the negative impact of GTBTs on the GTI of Chinese export enterprises.

Based on the methods of Liu and Ma (2020), this study builds Model (1) to examine the impact of GTBTs on the GTI of Chinese export enterprises (Liu and Ma, 2020). G T I i t + 1 c j = α 0 + β 1 G T B T j i t + θ X i t c j + μ i + λ j + γ c t + ε i t c j (1)

Furthermore, this study examined the impact of GTBTs and the LCCPP on the GTI of export enterprises. As the pilot work progressed, the National Development and Reform Commission issued pilot lists for low-carbon provinces and cities in 2012, selecting 92 cities included in the pilot scope as the experimental group, and the remaining cities as the control group. By constructing a multi-period DID model to evaluate the policy effect, exploring the response effect of the LCCPP on GTBTs and the effect of GTI, a model including interaction terms was constructed as follows: G T I i t + 1 cj = α 0 + β 1 G T B T i j t + β 2 T r e a t c × P o s t c t + β 3 T r e a t c × P o s t c t × G T B T i j t + θ X i t c j + μ i + λ j + γ c t + ε i t c j (2)

Where GTBT ijt is the core explanatory variable that represents the number of GTBTs encountered by the export enterprise i in the industry j of region c in the year t , measured by the logarithm of the environment-related TBT notifications affecting industry j in that year (Liu et al., 2023). GTI it + n cj represents the green technological innovation behavior of export enterprises in industry j of region c in year t+1, including green patent applications ( GTI − PAP it + 1 ) in year t+1 (Du et al., 2021; Shi et al., 2024) and green patent authorization ( GTI − PAU it + 1 ) in year t+1 (Peng et al., 2024; Peng et al., 2021). Post ct represents a time dummy variable, Treat c × Post ct represents a moderating dummy variable which is used to measure whether the export enterprises of industry j in region c are included in the low-carbon city pilot in year t (Chen et al., 2021; Chen et al., 2022; Song et al., 2020). X i t c j represents a series of control variables; μ i , λ j , γ c t represent the fixed effects at the enterprise level, fixed effects at the industry level, and fixed effects at the region year of the enterprise, respectively.

The main data sources include macro and micro data from the WTO Environmental Database, CNRDS, and CSMAR. This study uses Chinese A-share listed companies in Shanghai and Shenzhen from 2007 to 2021 as the research object to explore the impact of the GTBTs on the GTI of Chinese export enterprises. Considering the availability and validity of the data, the samples were screened and processed as follows. First, the data on GTBTs come from the WTO Environmental Database, which collects all the environment-related TBT notifications submitted by members to the WTO annually, including industry codes, countries submitting TBT notifications, HS6-digit codes, scope and targets of TBT application, and notification time. Second, using data from CSMAR, we sifted through the “sub-items” of “operating income” within the “profit and loss items” of each enterprise. We used the presence of the terms “foreign,” “foreign trade,” and “export” in any field to determine whether there was export behavior among A-share listed companies in Shanghai and Shenzhen from 2007 to 2021. We matched the HS6-digit codes and industry classification codes involved in the GTBTs from the WTO environmental database with the 2012 industry classification codes of the China Securities Regulatory Commission to obtain the industries of listed export enterprises affected by the GTBTs. Finally, financial listed companies, ST, ST*, PT enterprises, and enterprises with missing data related to the main research variables were excluded.

The data for measuring Chinese export enterprises’ GTI were obtained from the China Research Data Service Platform (CNRDS). Referring to Pan et al. (2022), this study identified the green patent data of listed export enterprises based on the “International Patent Classification Green List” launched by the World Intellectual Property Organization in 2010, combined with the International Patent Classification number, to obtain the annual green patent application and authorization number of enterprises, including green invention patents and green utility model patents (Pan et al., 2022). The quality of authorized green patents is higher than their quantity. The urban characteristic data and environmental indicators in this article are sourced from the “China Urban Statistical Yearbook.” All continuous variables were Winsorized at the 5% level to avoid the impact of outliers. Table 1 shows the definitions of all variables, and Table 2 presents the specific descriptive statistical results.

The regression results for the GTBTs and GTI of Chinese export enterprises are reported in Table 3, controlling for the fixed effects at the enterprise level, fixed effects at the industry level, and fixed effects at the region year of the enterprise. Columns (1) and (2) show the results without control variables, and columns (3) and (4) show the results after adding control variables. Considering both green patent applications and authorizations, the results show that the coefficient of GTBTs is negative and significant at the 1% level, regardless of whether the control variable is added, indicating that Chinese export enterprises experience a decline in both the number of green technology patent applications and authorizations when confronted with GTBTs. This indicates that the GTBTs negatively impact the GTI of Chinese export enterprises, thus verifying Hypothesis 1.

Table 4 shows the impact of the GTBTs on the GTI of Chinese export enterprises from the perspective of the moderating effect of the LCCPP. Columns (1) and (2) show the regression results, with the number of green technology patent applications and authorizations of enterprises in year t+1 as the explanatory variables. The impact coefficient of GTBTs on enterprises’ GTI is significantly negative, so when export enterprises suffer from GTBTs, the number of green technology patent applications and authorizations show a significant downward trend, which is consistent with Hypothesis 1. However, the interaction coefficient Treat c × Post ct × GTBT ijt is significantly positive, indicating that the number of green technology patent applications and authorizations of export enterprises supported by the LCCPP have decreased less, which has a significant promoting effect because the LCCPP increases the investment in enterprises’ green technology R&D, reduces their risk of GTI and speeding up the GTI output. Meanwhile, the increase in multi-policy support can alleviate GTBTs and financing constraints more effectively in export enterprises, making them focus on accelerating the application of patents for enterprises into the mass production of green products and improving the GTI performance of enterprises. Thus, Hypothesis two is verified.

If a few export enterprises in low-carbon cities enjoy substantial government support, this could create a crowding-out effect for those without such support, resulting in a misallocation of GTI resources and subsequently hindering the enhancement of export enterprises’ GTI capabilities. How can the intensity of a low-carbon city be quantified? First, we quantified the number of the LCCPP documents issued by governments between 2010 and 2021, with each city serving as a single unit. A higher number indicates a local government’s dynamic adjustment of green innovation goals and implementation of a precise policy supply, reflecting greater urban policy intensity. Consequently, the low-intensity group comprised cities with one to three policies, the medium-intensity group, four to six policies, the high-intensity group, seven or more policies, and the control group, with no policies. The generalized propensity score was estimated using a multinomial logit, yielding four propensity scores for cities categorized by intensity levels: 0, low, medium, and high. Subsequently, the reciprocal of each propensity score was calculated to generate the sampling weights (Guo and Fraser, 2015). e (Xa,b) = pr (B = b |X = x) is the generalized tendency value of sample a under the influence of policy strength b, X represents the observed covariate, and 1/e (Xa,b) denotes the sampling weight of sample a under the influence of policy strength b. Regression analysis is employed to assess the impact of policy strength. Based on the cities categorized by intensity levels, the regression of Equation 2 was carried out in groups, and the results are shown in Table 9.

When policy intensity is low, the interaction coefficient Treat c × Post ct × GTBT ijt is positive but not significant, suggesting that the moderating effect is not significant and that it is difficult to stimulate the GTI of export enterprises. At moderate policy intensity, the interaction coefficient Treat c × Post ct × GTBT ijt is positive and significant at the 5% level. The results indicate that the LCCPP of medium intensity mitigates the negative effects of GTBTs on export enterprises’ GTIs. More importantly, the moderating effect of the LCCPP is significant under high intensity, and the interaction coefficient is significant at the 1% level, indicating that these pilot policies at high levels will help export enterprises effectively resist the negative impact of GTBTs on GTI. The possible explanations are provided for this finding, at moderate policy intensity, although moderate policy intensity can alleviate the GTI of GTBTs to some extent, this moderating effect is limited to the upper limit of administrative penalties or incentive support. When the compensation benefits obtained from green innovation activities by enterprises are equivalent to the costs caused by GTBTs, and cannot obtain more compensation from the market, moderate policy intensity is also difficult to continuously improve the GTI of export enterprises. On the contrary, When the negative effects caused by GTBTs and the penalty costs caused by environmental non-compliance are covered by the innovation compensation effect of LCCPP at a high intensity, it provides sustainable conditions for the GTI of enterprises, forming the so-called “Porter hypothesis”.

The implementation of the LCCPP encompasses various policy instruments that may differently influence the moderating effect of GTBTs on export enterprises’ GTIs. Referring to the classification of Du et al. (2021), this study divides policy instruments into three categories (Du et al., 2021). First, command-based policy instruments mainly restrict enterprise emissions by formulating strict emission reduction targets and green technical standards, which will significantly achieve cleaner production standards through the transformation of production equipment, but also increase the cost of pollution control for enterprises, forcing enterprises to choose the path of technological transformation to achieve green transformation. Market-based policy instruments comprise a range of incentive policies, including carbon trading mechanisms, clean development mechanisms, subsidies for green innovation, incentives for talent, green finance and funds, and tax preferences. These policies guide enterprises to conduct green technology R&D in accordance with the latest technical regulations and standards, reduce R&D risks, and precisely incentivize enterprises to achieve green transformation using market-oriented approaches, such as pricing, subsidies, and taxation. Third, voluntary policy instruments like implementing the LCCPP, including the construction of a low-carbon transportation system and preparing for low-carbon industrial parks, will encourage enterprises’ spontaneous environmental protection behavior through publicity.

The performance of the GTI by export enterprises in response to the GTBTs, with the moderating effect of the three types of the LCCPP instruments, is presented in Table 10. The results show that only market-based policy instruments have a significant moderating effect and that the interaction coefficient Treat c × Post ct × GTBT ijt is positive and significant at the 1% level, indicating that market-based policy instruments alleviate the negative effects of GTBTs on exporting enterprises’ GTI. However, the moderating effect of command-based policy instruments is not significant, and voluntary policy instruments exacerbate the negative effects of the GTBTs on the GTI of exporting enterprises. One possible reason is that although command-based policy instruments can strictly restrict the emission reduction behavior of polluting enterprises, but have less effective in spurring GTI, enterprises will make green innovation decisions in the dynamic balance between economic performance and environmental performance (Pan et al., 2022). When some export enterprises are unable to break through the constraints of GTBTs and environmental regulations, as a result, exiting the market or trade transfer becomes common strategies. For enterprises with certain financial and technological foundations, most enterprises will replace GTI with technological transformation to achieve the objectives of command-based policies. On the one hand, enterprises will enhance end-of-pipe treatment to quickly meet the standards set by environmental regulations, and on the other hand, they will achieve clean production standards by purchasing pollution control equipment and green production equipment, green technology transformation is more realistic and feasible than GTI (Chen et al., 2022). Voluntary policy instruments may have an extrusion effect on profitable project investments, increase enterprise costs, hinder improvements in enterprise productivity, and reduce profits. Voluntary policy instruments will not create an “innovation compensation” effect, and will also further aggravate the negative impact of GTBTs on GTI of export enterprises.