Introduction

The growing threat of global warming, driven primarily by massive carbon emissions, endangers human well-being and sustainable development worldwide. Due to the negative externalities of environmental issues, urgent global action is needed from governments to curb the escalating trend of carbon emission intensity. As the rapid expansion of the secondary industry has made China the largest contributor to global carbon emissions, controlling China’s surging carbon emissions has become a topic of intense debate. In 2020, the Chinese government introduced the “double carbon” targets, with the aim of reaching the peak of carbon emissions by 2030 and achieving carbon neutrality by 2060. To achieve these targets, the carbon emission intensity must be dramatically reduced, requiring a decrease in carbon emissions per unit of GDP by 60 to 65% by 2030 compared to 2005 (Zhu and Lee, 2022). This highlights the importance of understanding the factors affecting carbon emission intensity.

The topic of reducing carbon emission intensity has been thoroughly explored in existing literature, which has identified multiple factors contributing to this process. One crucial driving factor is the upgrade of industrial structure, where the ratio of value added by the tertiary industry to the gross domestic product is considered a promising way to reduce carbon emission intensity due to the less resource-intensive nature of modern service sectors (Zhang et al., 2014). For example, Zhang & Ma (2020) analyzed 21 industrial sectors in eight developed countries, finding that transitioning to clean service industries can effectively reduce carbon emission intensity. Furthermore, technological progress also plays a critical role in reducing carbon emission intensity by improving economic efficiency (Cheng et al., 2018; Zhang et al., 2020). Other factors that have been found to impact carbon emission intensity include research and development investment (Huang, 2021; Huang et al., 2020, 2018), and trade openness (Wang, 2021). Given the indispensable role of government in environmental governance, numerous studies have explored the effectiveness of environmental regulation on reducing carbon emissions (Guo and Wang, 2018; Pei et al., 2019; Wu et al., 2020; Yang et al., 2020). Similarly, carbon emission trading pilot projects and low-carbon city pilot policies implemented by the Chinese government have also been shown to contribute to carbon emissions reduction (Liu et al., 2022; Shi et al., 2022; Wang et al., 2024; Zhang et al., 2020). The above literature is primarily based on the theory of public goods, suggesting that the government should regulate environmental issues and explore various effective governmental measures in reducing carbon emissions. Alternatively, other studies seek to identify positive factors from the supply side that contribute to carbon reduction. Some studies related to the present topic have highlighted the importance of the service sector and consumption demand in reducing carbon emission intensity (Zhou et al., 2020) or have demonstrated that the tertiary industry is a relatively low-carbon sector (Wang et al., 2024; Yi et al., 2016).

While there is a growing body of literature exploring approaches to reduce carbon emission intensity (Shobande et al., 2024; Zeng et al., 2023), some research gaps still exist. First, most prior research on reducing carbon emissions focuses predominantly on supply-side factors such as technological advancements and environmental regulations rather than demand-side determinants (Ai et al., 2023; Shao et al., 2021; Wu et al., 2020). However, shifting consumer preferences toward low-carbon goods and services can encourage firms to produce environmentally friendly offerings, contributing significantly to carbon emission intensity reduction. However, few studies have focused on demand-side factors. Second, although cultural products constitute an important component of the tertiary industry (Zhao and Liu, 2020), few studies have examined their role in reducing carbon emission intensity. The consumption of cultural products, such as media and entertainment, publishing, visual arts, heritage and museums, crafts and design, may not only have the potential to directly reduce carbon emissions at the production side, but also contribute to carbon reduction through the promotion of environmental awareness. To address these gaps, the aim of the current study is to empirically examine the impact of one novel cultural consumption policy (CCP) on carbon emission intensity.

China provides a unique context for studying this issue. A rapid increase in resident income has been observed over the past four decades, leading to an expectation of a preference for cultural goods. Unlike products from the secondary industry, cultural goods and services belong to the tertiary industry, which comprises services such as tourism and cultural activities and is less carbon-intensive. Furthermore, the consumption of cultural goods, such as musical performances and cultural tourism, provides emotional and spiritual fulfillment to consumers without requiring direct inputs of natural resources (Deng et al., 2019).

Utilizing panel data of 280 Chinese prefecture-level cities from 2008 to 2019, this research seeks to explore the causal effect of the CCP in China on carbon emission intensity with a difference-in-differences (DID) model. This study contributes to relevant literature in three aspects. First, based on the examination of the carbon emission intensity reduction effects of the CCP, we find that promoting cultural consumption helps to lower carbon emission intensity. We offer complementary evidence on the question of how to promote carbon emission intensity reduction. To our knowledge, this is the first paper that empirically explores carbon emission intensity reduction from the perspective of cultural consumption. Second, when examining how the CCP can facilitate the reduction of carbon emission intensity, we identify that industrial upgrading, changes in employment structure, and increases in resident consumption as potential mechanisms through which the CCP operates. Third, this study enhances understanding of cultural consumption’s role in supporting China’s “double carbon” targets, promoting the development of cultural consumption and the cultural industry through supportive policies can contribute to achieving the goal of carbon emission reduction.

The rest of the paper is organized as follows. Section “Policy background and theoretical framework” presents policy background and theoretical framework. Section “Empirical methodology and data” describes empirical methods and data. Section “Results” reports our main empirical findings, including benchmark regression results, robustness tests, and mechanism analysis. Followed by the discussion in Sections “Discussion” and “Conclusions and policy implications” provides the conclusions and policy implications.

Policy background and theoretical framework

In China, the cultural industry has become a significant driver of economic growth (Zhou et al., 2020). To foster the development of cultural industry and encourage cultural consumption, the former Ministry of Culture and the Ministry of Finance jointly initiated a pilot project aimed at stimulating resident cultural consumption in 2015. Following this initiative, in 2016, 26 cities were designated as the first batch of national cultural consumption pilot cities. This marked a significant shift in national policy priorities, emphasizing cultural consumption as a strategic focus.

These pilot cities implemented various policies tailored to local social and economic conditions to promote cultural industry growth and stimulate cultural consumption (see Table 1). These policies can be categorized into three primary types. First, cities focused on creating diverse cultural consumption opportunities, actively hosting cultural and art festivals, and organizing music and dance performances. For example, Ningbo expanded cultural consumption opportunities by hosting festivals such as the East Asian Cultural Capital 2016 Ningbo Year Series, Zhejiang Choral Festival, Ningbo International City Art Fair, and the 3rd Citizen Culture and Art Festival. Additionally, the Ningbo government encouraged private investment in cultural consumption activities. Similarly, the Yinchuan Government introduced policies like the “Measures for Supporting Key Literary and Art Projects in Yinchuan” and the “Measures for Rewarding Literary and Artistic Works Awarded by National and Autonomous Regions in Yinchuan.” allocating 15 million yuan annually for outstanding works and 5 million yuan for internet movies, cultural creativity, and digital design services. Second, some cities provided cultural consumption subsidies. For instance, Hefei allocated 8 million yuan in 2017 to subsidize consumers’ cultural consumption expenditures, with a subsidy ratio of 50% for theater performances, including dramas, operas, and symphonies. Third, some other cities implemented policies aimed at enhancing cultural consumption participation through incentive measures. Wuhan, for example, established the WeChat public account “E Wen Tianxia”, which integrated cultural consumption venues, consumers, cultural enterprises, and government departments. Residents could earn consumption points and coupons by rating cultural goods. In summary, pilot cities employed a range of policies to diversify cultural consumption opportunities and reduce the cost of resident cultural consumption.

Table 1 Summary of the cultural consumption policy.
Full size table

The implementation of the CCP can contribute to reducing carbon emissions through three distinct pathways, as depicted in Fig. 1. First, as resident income increases, there is a shift in consumption patterns, with more expenditures allocated to cultural goods (Brito and Barros, 2005). The CCP, with its cultural consumption subsidies, further stimulates demand for cultural goods, including entertainment, sports, fitness, and tourism. These are types of consumption involving non-tradable goods, which can only be consumed locally. Moreover, such consumption activities, representing an upgrade in consumption patterns, are characterized by being environmentally friendly; some forms of cultural consumption are even completely zero-emission. For example, museum visiting is a sightseeing experience that brings spiritual pleasure without directly consuming industrial products that require fossil fuels in their production processes.

Fig. 1
figure 1

Theoretical framework.

Full size image

Second, financial support for the cultural industry has facilitated a shift in employment structure. Fiscal assistance to cultural enterprises has reduced their operational costs. Additionally, subsidies for cultural consumption have lowered the prices of cultural products, stimulating an increase in demand for such products. This surge in demand drives the expansion of cultural enterprises, which in turn absorb more workers into the cultural sector. Consequently, these policies promote employment in a sector characterized by lower carbon emission intensity compared to traditional industries. As an example, in 2018, Shenzhen’s cultural industry contributed over 190 billion yuan to the city’s GDP, accounting for 7.9% of its total output, the proportion of Shenzhen’s cultural and creative industry to Shenzhen’s GDP has exceeded 10% for the first time in this year. By 2019, Shenzhen boasted more than 50,000 cultural enterprises, employing nearly 1 million individuals. Furthermore, Shenzhen’s cultural industry park hosted over 8000 enterprises with a total revenue exceeding 150 billion yuan.

Third, the CCP can contribute to the reduction of carbon emission intensity by facilitating industrial structure upgrades. Increased cultural consumption and the growth of the cultural industry’s labor force accelerate the development of the tertiary sector, particularly service-based industries. This shift toward a service-oriented industrial structure, driven by the CCP, holds significant potential for reducing carbon emission intensity. For instance, the production of cultural goods typically involves lower energy consumption due to their emphasis on spiritual enjoyment and reduced reliance on fossil fuels.

Empirical methodology and data

Empirical methodology

To estimate the effect of the CCP on carbon emission intensity, we applied a difference-in-differences (DID) regression design:

$${\rm{ln}}\,{C}{I}_{it}={\alpha }_{0}+{\alpha}_{1}{CCP}_{it}+\sum {\alpha }_{j}{Control}_{it}+{\delta }_{i}+{\eta }_{t}+{\varepsilon}_{it}$$
(1)

where lnCIit stands for the logarithm of carbon emission intensity in city i at year t, CCPit is the key independent variable, representing a city’s CCP status. Specifically, CCPit=treat*post, where treat = 1 if cityi carried out the CCP during the sample periods, treat = 0 otherwise, Post is a post-treatment indicator, takes a value of 1 if Post\(\ge\)2016, takes a value of 0 else. Controlit is a vector of control variables, including industry structure, population, R&D, and openness. δi denotes city fixed effects, ηt denotes year fixed effects, εit is the error term. A key assumption of the DID regression is the parallel trend, which means that the treatment group and the control group should exhibit the same trend in outcomes in the absence of the policy intervention. If the trends are the same, the parallel trend assumption is satisfied, validating the effectiveness of the DID regression. Therefore, to examine whether the impact of the CCP is robust, we will conduct a parallel trends test after the DID regression.

Data and variable definitions

Explained variable

The explained variable in this study is carbon emission intensity, calculated as the ratio of carbon emissions to GDP. However, statistical data do not directly provide carbon emission calculations in Chinese prefecture-level cities. Following the methodology described in Wang and Zhang (2022), we calculate a city’s carbon emissions by analyzing its consumption of electricity, natural gas, and liquefied petroleum gas. This approach is considered reasonable given China’s distinct energy structure, characterized by an abundance of coal, limited oil reserves, and relatively low levels of natural gas (Liu et al. 2021). Coal-fired power generation, in particular, holds a predominant position in China’s energy landscape and contributes significantly to carbon dioxide emissions. This methodology allows us to estimate carbon emissions in the absence of direct statistical data, providing a valuable basis for assessing carbon emission intensity in Chinese prefecture-level cities. Finally, we calculate carbon emission intensity by dividing carbon emissions by GDP. We employ Eq. (2) to calculate carbon emissions, displayed as follows.

$$C{O}_{2}={C}_{{\rm{p}}}+{C}_{n}+{C}_{e}=k{E}_{p}+v{E}_{n}+\beta (\eta \times {\lambda }_{e})$$
(2)

In this equation, Cp, Cn, and Ce represent carbon emissions attributed to liquefied petroleum gas, natural gas, and electricity consumption, respectively. Ep corresponds to liquefied petroleum gas consumption, while En denotes natural gas consumption, k and v are the carbon dioxide emission coefficients associated with liquefied petroleum gas and natural gas, respectively, and β signifies the carbon dioxide emission coefficients for coal-fired power generation (See Table 2).

Table 2 CO2 emission coefficient.
Full size table

λe denotes electricity consumption, and η represents the share of coal power generation in the total power generation, as outlined in Table 3.

Table 3 Proportion of coal power generation (%).
Full size table

Table 4 presents the growth rate of carbon emissions in the sample cities, as well as the growth rates before and after the implementation of the CCP. As evidenced by the data, carbon emissions increased by 170.6% across all sampled cities, with a 44.3% increase before the implementation of the CCP and a 78.2% increase after. We further divided the sample based on whether a city was a CCP pilot city. In pilot cities, carbon emissions increased by 92.5%, while in non-pilot cities, the increase was even greater at 214.5%. This difference became particularly pronounced after the implementation of the CCP, with pilot cities experiencing a 33.7% increase in carbon emissions, significantly lower than the 104.3% growth rate in non-pilot cities. This reveals that after the implementation of the CCP, the growth rate of carbon emissions in pilot cities significantly slowed down.

Table 4 Carbon emissions growth rate before and after the CCP implementation.
Full size table

Explanatory variable

The CCP is the core independent variable in this study, indicating whether a city is a pilot city for cultural consumption policies. Specifically, the CCPit is defined as the interaction term of Treat and Post, where Treat equals 1 if city i is a pilot city during the sample period and 0 otherwise. Meanwhile, the variable Post serves as a post-treatment indicator, taking a value of 0 if the year is before 2016 and 1 if the year is 2016 or later, corresponding to the year of the CCP implementation.

Control variables

Several variables that may influence carbon emission intensity are incorporated as control variables. (1) Industrial Structure: Measured by the proportion of the secondary industry’s output to GDP. The secondary industry, compared to both primary and tertiary industries, exhibits the most substantial CO2 emissions. Therefore, it bears a direct relationship with a city’s carbon emission intensity. (2) Population: We employ the logarithm of the registered population count, as it plays a pivotal role in influencing energy consumption patterns. (3) Income (Salary): Average income is measured in logarithmic form, as it is expected that the demand for a cleaner environment increases as income levels rise. (4) Research and Development (R&D): Following the methodology of previous studies (Pan et al., 2021), we use government scientific expenditure as a proxy for R&D. R&D is believed to enhance innovation capabilities and economic efficiency, consequently fostering reductions in carbon emission intensity. (5) Openness: Assessed by the logarithm of the amount of foreign capital utilized in a given year relative to GDP. We convert foreign investment values from US dollars into RMB units using yearly exchange rates. Openness can either lead to a pollution-halo effect, restraining carbon emission intensity, or a pollution-haven effect, promoting it.

Mechanism variables

To investigate the possible channels through which the CCP affects carbon emission intensity reduction, we considered the following influencing mechanisms. The mechanism variables employed in this study encompass: (1) Industrial Upgrade: Measured as the ratio of tertiary industry output to secondary industry output. This metric captures the extent of industrial transformation and modernization. (2) Secondary Industry Employment: This variable represents the proportion of secondary industry employment in relation to total employment figures. It sheds light on the labor distribution within the secondary sector. (3) Tertiary Industry Employment: Measured as the ratio of tertiary industry employment to total employment. This ratio reflects the composition of the workforce in the tertiary sector. (4) Resident Consumption: Calculated by dividing the total retail sales of social consumer goods by population. This indicator offers insights into the level of resident consumption.

Data sources

Table 5 provides the descriptive statistics of the sample, which encompasses 280 Chinese prefecture-level cities from 2008 to 2019. The starting point of our sample is 2008 because the statistical information before 2008 was not sufficiently complete, and there were many missing values in the data. To ensure the accuracy of our empirical analysis, we used data from 2008 onward. The reason for choosing 2019 as the end point of our sample is that COVID began to spread in early 2020, and the Chinese government implemented relatively extreme lockdown policies, which significantly curtailed many economic activities. This has a major impact on the dependent variable in this study, carbon emission intensity. Therefore, we did not include data after 2019 to ensure that our empirical research is not affected by COVID and lockdown policies. Importantly, our explanatory variable, the implementation of the CCP, occurred in 2016. Using data from 2008 to 2019 is sufficient to evaluate the short-term effects of the CCP. The primary data sources for this study include the China Urban Statistical Yearbook, which provides essential variables encompassing energy consumption and control factors. It is worth noting that ten cities had to be excluded from our analysis due to insufficient data availability. Information pertaining to the proportion of coal-fired power generation was extracted from the China Electricity Yearbook, while carbon emission coefficients were sourced from the Intergovernmental Panel on Climate Change (IPCC).

Table 5 Descriptive statistics.
Full size table

Results

Benchmark regression results

Table 6 presents the benchmark regression results, with control variables introduced incrementally. In Column (1), we report the results without any control variables. Notably, the coefficient for the CCP is −0.3763 and statistically significant at the 1% level, indicating a significant reduction in carbon emission intensity. As we proceed to incorporate additional control variables, we observe some slight variations in the effect of the CCP. In Column (2), when we include industrial structure as a control, the magnitude of the CCP effect diminishes somewhat but remains statistically significant.

Table 6 Benchmark regression results.
Full size table

Columns (3) and (4) subsequently incorporate population and salary as control variables. Population does not demonstrate a statistically significant impact on carbon emission intensity, while an increase in salary is correlated with a reduction in carbon emission intensity. Columns (5) and (6) incorporate R&D and openness into the regression model, and neither of these variables exhibits a significant effect. Importantly, in Column (6) where all control variables are included, the CCP coefficient remains significant, measuring at −0.3229 and significant at the 1% level. Overall, the results in Table 6 suggest that the implementation of the CCP has a statistically significant effect on reducing carbon emission intensity.

Robustness tests

To ensure the reliability and robustness of our estimates, we perform a series of robustness tests in this section.

Parallel trend test: event study

Crucially, the validity of DID estimates depends on satisfying the parallel trend assumption. This assumption requires that, before the implementation of CCP, the trend in carbon emission intensity should be comparable between pilot cities and non-pilot cities. To investigate this issue, we employ an event study approach using the following equation.

$${\mathrm{ln}}\,C{I}_{it}={\alpha }_{0}+{\alpha }_{1}CC{P}_{it}+\mathop{\sum }\limits_{k=-6}^{3}{\beta }_{k}\times {D}_{i,{t}_{o+k}}+{\delta }_{i}+{\eta }_{t}+{\varepsilon }_{it}$$
(3)

where Di,t0+k represents a series of dummy variables, with k indicating the time relative to the year of the CCP implementation. Specifically, t0 denotes the year when the CCP was enacted in city i, and k represents the kth year relative to the CCP implementation year. The timeframe of analysis spans six years preceding the CCP to three years following its implementation. The core coefficient, βk is pivotal as it captures the differences in carbon emission intensity between pilot cities and other cities. The parallel trend assumption is satisfied when βk exhibits a smooth curve with k \(<\) 0, signifying consistency with the parallel trend hypothesis. This consistency implies that there were no significant differences in carbon emission intensity between pilot cities and non-pilot cities prior to the implementation of the CCP.

Figure 2 visually presents the results of the parallel trend test. It is evident that the coefficient exhibits a gradual curve when k < 0, signifying that carbon emission intensity in both pilot and non-pilot cities did not significantly differ before the CCP was enacted. However, when k = 1, the coefficient experiences a substantial decline and maintains a consistently negative value in subsequent periods. Therefore, the parallel trend test demonstrates that there was no significant difference in carbon emission intensity between the treatment group and the control group before the implementation of the CCP. After the implementation of the CCP, differences began to emerge between the two groups, providing evidence for the effectiveness of this policy.

Fig. 2
figure 2

Parallel trend test.

Full size image

Excluding other overlapping policies

Our benchmark regression analysis demonstrates the role of the CCP in reducing carbon emission intensity. However, other carbon-reduction policies in China, including low-carbon city pilots and carbon trading pilots, may confound these results. The Chinese government initiated low-carbon city pilots in 2010, 2012, and 2017, aiming to establish low-carbon industrial systems, lifestyles, and consumption patterns while reducing carbon emission intensity. Additionally, the carbon trading pilot policy, introduced in 2013, directly regulates carbon emissions in nine pilot provinces or cities. Prior research has demonstrated the effectiveness of these policies in curbing carbon dioxide emissions (Liu et al., 2022; Shen et al., 2018; Zhang et al., 2019).

To address the potential influence of these other policy tools on our estimates, we have excluded cities that participated in the low-carbon pilot programs in 2010 and 2012, as well as the nine provinces and cities involved in the carbon trading pilot program. Cities selected for low-carbon pilots in 2017 are unaffected, as their implementation post-dates the CCP in 2016. After excluding these potentially interfering samples, we conduct a DID regression. Three potential outcomes arise: if the coefficient is significant and its magnitude increases, it suggests that our benchmark estimate is likely underestimated, demonstrating the robustness of our results. Conversely, if the coefficient is significant and its magnitude decreases, it implies that our benchmark estimate may be overestimated, but the results remain robust. If the coefficient is insignificant, it indicates that our benchmark estimates are unreliable.

The results are reported in Table 7. Columns (1) and (2) present the results excluding low-carbon pilot cities. Column (1), without controls, shows a significant negative coefficient. The value of the coefficient is smaller compared to the result in the benchmark regression presented in Column (1) of Table 6. When controls are incorporated into the regression, the coefficient value is −0.2271, which is still smaller but significant compared to the benchmark regression. Thus, excluding the low-carbon pilot cities slightly overestimates the effect of the CCP on carbon emission intensity, but it is still significant. Columns (3) and (4) show similar results when excluding carbon trading pilot cities: the DID coefficient is smaller but statistically significant without controls, and including controls confirms that the CCP remains significant in reducing carbon emission intensity. In conclusion, the results suggest that our benchmark regression results are robust, and the CCP remains a crucial factor in reducing carbon emission intensity, despite a slight overestimation.

Table 7 Robustness checks: excluding other confounding policies.
Full size table

Placebo test

Another concern within the context of the DID model is the potential impact of unobserved, time-varying variables on the outcome. Despite our efforts to control for specific observed factors such as GDP, industrial structure, population, salary, R&D, and openness level, it is not feasible to account for all variables that may influence carbon emission intensity, especially those that are unobserved.

To address this concern, we have conducted a placebo test, a well-established and robust assessment within DID models. We undertook two key steps. First, we randomly selected 26 cities as placebo “pilot cities”, matching the number of actual CCP pilot cities, and estimated a DID regression to assess the placebo effect on carbon emission intensity. Second, we repeated this process 500 times, each time generating a false estimate based on a new set of randomly selected cities.

Figure 3 illustrates the results of the placebo test. The curve in the figure illustrates the distribution of estimated coefficients derived from the randomly selected samples. These coefficients tend to cluster around zero and exhibit a normal distribution. The red vertical line on the graph represents the outcome of our benchmark regression. Significant discrepancies emerge between the benchmark regression results and the estimated coefficients of the randomly selected samples. This disparity confirms the statistical robustness of our findings, reinforcing the CCP’s significant effect on carbon emission intensity reduction.

Fig. 3
figure 3

Placebo test.

Full size image

Underlying mechanism identification

In this section, we explore the underlying mechanisms through which the CCP can exert an influence on carbon emission intensity, encompassing the upgrading of industrial structure, improvements in employment composition, and shifts in consumer expenditure. The results are presented in Table 8. In Column (1), we assess the impact of the CCP on industrial upgrade, quantified by the ratio of tertiary industry output to secondary industry output. The estimated coefficient for the effect of the CCP on industrial upgrade is positive and statistically significant at a 5% level. This finding underscores the pivotal role of industrial upgrade in driving the reduction of carbon emission intensity as a result of the CCP. Furthermore, the secondary industry accounts for the majority of coal-based energy consumption in China, where the energy profile is characterized by abundant coal, limited oil, and scarce natural gas. A decrease in the proportion of the secondary industry not only facilitates a reduction in high-carbon energy consumption but also signifies a positive shift toward lower fossil energy utilization within the tertiary sector.

Table 8 Mechanism test.
Full size table

Our analysis continues by examining changes in the employment structure, as presented in Columns (2) and (3) of Table 8. In Column (2), we explore the impact of the CCP on secondary industry employment, revealing a coefficient of −3.3330, which is statistically significant at the 1% level. This suggests that the implementation of the CCP has led to a reduction in employment within the secondary industry. Conversely, Column (3) presents a positive and statistically significant coefficient at the 1% level, indicating a notable increase in employment within the tertiary industry.

Given that one of the aims of the CCP is to stimulate cultural and tourism consumption, we extend our analysis to investigate resident consumption in Column (4) of Table 8. Our findings reveal a coefficient of 1.1862, significant at the 1% level, indicating an increase in resident consumption as a result of the CCP. This boost in resident consumption plays a pivotal role in driving the development of the tertiary industry, ultimately contributing to the industrial structure’s upgrading and a consequent reduction in carbon emission intensity.

Discussion

For decades, China’s remarkable economic growth, often referred to as the “China Miracle,” has been driven by industrial development. However, this development pattern has come at a substantial cost, including high carbon emission intensity and environmental pollution. In response to these environmental challenges, the Chinese government announced ambitious “double carbon” targets in 2020, with the goal of reaching peak carbon emissions by 2030 and achieving carbon neutrality by 2060. Hence, it is crucial to identify effective strategies to reduce carbon emission intensity in order to achieve these targets.

From a demand-side perspective, this study explores the impact of the CCP on carbon emission intensity, utilizing data from 280 prefecture-level cities across China. Our findings provide compelling evidence that the CCP significantly reduces carbon emission intensity. While previous research has predominantly focused on supply-side factors such as technological progress, research and development (R&D), and innovation (Cheng et al., 2018; Huang, 2021; Huang et al., 2020; Zhang & Ma, 2020), this study offers a novel perspective on cultural demand by examining the role of the CCP. Our findings emphasize the importance of demand-side factors and the effective role of the cultural industry in reducing carbon emission intensity.

We explore the mechanisms by which the CCP contributes to reducing carbon emission intensity. The CCP significantly accelerates the upgrading of the industrial structure, consistent with prior research by Zhang & Gu (2022). Their study, which employed panel data from 266 Chinese prefecture-level cities, affirmed the CCP’s pivotal role in promoting industrial structure upgrading. The secondary industry primarily encompasses manufacturing, construction, and other sectors related to the production process. The tertiary industry mainly refers to service sectors such as cultural activities, tourism, and retail, contribute to reduced carbon emission intensity by prioritizing low-carbon activities. A large amount of carbon emissions originates from the production process, which means that the carbon emissions from the secondary industry are much higher than those from the tertiary industry. The increase in the proportion of the tertiary industry will lead to a reduction in carbon emission intensity (Zhang et al., 2023). This aligns with a broader body of literature emphasizing the role of industrial structure upgrades in mitigating carbon emission intensity. For instance, Zhang et al. (2014) emphasized that the development of modern service sectors, such as the cultural industry with its lower resource consumption, can significantly reduce carbon emission intensity.

Another mechanism is the shift in the employment structure ratio, with the CCP leading to an expansion of employment in the tertiary sector and a concurrent reduction in the secondary sector. Specifically, the increase in employment in the cultural industry has promoted the expansion of both the cultural sector and the tertiary industry. The expansion of these low-carbon industries helps to decrease energy consumption and reduce the overall carbon emission intensity of the economy. This enhancement is consistent with the policies implemented by the CCP pilot cities to strengthen the cultural industry, such as cultural consumption subsidies in Hefei and dedicated funds to promote cultural industry growth in Yinchuan. These policy measures have led to increased job opportunities and higher consumption within the tertiary industry, thereby bolstering the development of a low-carbon economy.

Following industrial structure and employment shifts, the third mechanism involves a significant increase in resident consumption. In contrast to industrial products, cultural commodities often exhibit characteristics of immediate consumption and non-tradable (Deng et al., 2019). The increase in local consumption levels reflects the effectiveness of cultural consumption policies. This type of increase in environmentally friendly consumption contributes to the green and low-carbon effect of this industry and helps to reduce carbon emission intensity. This also serves as evidence of the CCP’s objectives aligning with its successful implementation.

Nonetheless, it is essential to acknowledge several limitations in this study. First, the collection of data on Chinese prefecture-level cities is subject to constraints, limiting our capacity to analyze the heterogeneous effects of the CCP across diverse cultural industry sectors. Second, the implementation of the CCP may exert a substantial influence on the behavior of market entities, especially cultural enterprises. Third, in the supply of cultural products, digital technologies and other AI tools are increasingly being used. The application of these technologies may consume computational power and generate a certain level of carbon emissions. Future research using enterprise-level data could provide deeper insights into the CCP’s effects on cultural enterprises and their behavior, as well as the carbon emissions generated by the application of AI in cultural products. Moreover, as this study focuses on China, the generalizability of our findings to other contexts remains unclear. The implementation of cultural consumption pilot programs in certain Chinese cities provides a suitable context for our empirical assessment; however, this context may not necessarily be universally applicable internationally. Whether policies for developing cultural industries and stimulating cultural consumption in different countries have a carbon reduction effect calls for further exploration.

Conclusions and policy implications

This research offers a fresh perspective on reducing carbon emission intensity. Using data from Chinese prefecture-level cities, our study employs the DID design to demonstrate the effective role of the CCP in reducing carbon emission intensity. The robustness of our findings is confirmed through a series of rigorous tests, highlighting the potential of policies that bolster cultural consumption in achieving carbon emission intensity reduction objectives. The beneficial role of the CCP in lowering carbon emission intensity underscores the significance of fostering cultural consumption and nurturing the cultural industry within the framework of China’s “double carbon” targets. Currently, China’s cultural consumption pilot policy is expanding its purview to encompass diverse industrial sectors, including the tourism industry. It is reasonable to expect that the continued expansion and refinement of the CCP will assume a pivotal role in the realization of China’s objectives to attain carbon peak by 2030 and carbon neutrality by 2060.

The implications of our findings hold significant policy relevance for China as well as other emerging economies in transitional development phases. Given the urgency of achieving the “double carbon” targets, demand-side measures should be prioritized to drive carbon emission intensity reduction. Policies aimed at shifting consumption patterns, including financial incentives and consumption subsidies to reduce the cost of low-carbon cultural products, should be actively promoted. Furthermore, policymakers should fully recognize the pivotal role of industrial upgrades, with a special emphasis on the contribution of the cultural industry to carbon emission intensity reduction. To further decarbonize the economy, local governments can leverage modern digital technology and promote financial subsidies to enhance the cultural product consumption experience and cultivate a diverse cultural consumption ecosystem, thereby enhancing the allure of cultural products.