In recent years, the electric vehicle sector in China has experienced remarkable growth, positioning the country as a global leader in the new energy automotive market. According to industry reports, China’s electric vehicle production has surged from a modest 18,000 units in 2013 to over 9 million units by 2023, capturing more than 60% of the global market share. This rapid expansion underscores the critical role of technological innovation and strategic alliances in driving the industry forward. As a researcher focused on innovation dynamics, I aim to explore how technology absorption capacity within innovation alliances influences the innovation performance of electric vehicle enterprises in China. Specifically, I investigate the interplay between market competition and government subsidies, which are pivotal in shaping the outcomes of such collaborations. The electric vehicle industry in China, often referred to as China EV, represents a key segment of the broader push toward sustainable transportation, making it an ideal context for this study.
The concept of innovation alliances has gained traction as a means for electric vehicle companies to pool resources, share knowledge, and accelerate technological advancements. These alliances, such as the Electric Vehicle Industry Technology Innovation Strategic Alliance in China, provide a platform for firms to access external technologies and collaborate on research and development. However, the mere participation in such alliances does not guarantee improved innovation performance; instead, it hinges on a firm’s ability to absorb and internalize the acquired technologies. Technology absorption capacity, defined as the capability to identify, assimilate, and apply external knowledge, is thus a critical determinant of success in the electric vehicle sector. In this paper, I delve into the mechanisms through which technology absorption capacity impacts innovation performance, while considering the moderating role of market competition and the mediating effect of government subsidies. The electric vehicle market in China, or China EV, serves as a dynamic backdrop for this analysis, highlighting the unique challenges and opportunities faced by firms in this rapidly evolving industry.
To structure this investigation, I begin by reviewing existing literature on technology absorption capacity, innovation performance, and the specific context of electric vehicles in China. This leads to the formulation of hypotheses that guide the empirical analysis. Subsequently, I describe the research design, including data sources, variable definitions, and econometric models. The empirical results are presented through descriptive statistics, correlation analyses, and regression outputs, followed by robustness checks and heterogeneity analyses based on ownership types and regional distributions. Finally, I explore the mechanisms of market competition and government subsidies, concluding with practical recommendations for electric vehicle enterprises in China. Throughout this paper, I emphasize the importance of technology absorption capacity as a driver of innovation in the China EV landscape, underscoring how firms can leverage alliances to enhance their competitive edge.
Literature Review and Hypotheses Development
The relationship between technology absorption capacity and innovation performance has been extensively studied in various industries, but its application to the electric vehicle sector in China warrants deeper examination. Drawing on resource-based theory and synergistic effects, technology absorption capacity enables firms to transform external knowledge into internal competencies, thereby fostering innovation. In the context of China’s electric vehicle industry, or China EV, this capacity is particularly crucial due to the high pace of technological change and the need for continuous innovation in areas like battery technology, autonomous driving, and energy efficiency. Prior research, such as that by Cohen and Levinthal (1990), established that absorption capacity facilitates learning and innovation by allowing firms to effectively utilize external information. For electric vehicle companies in China, participation in innovation alliances provides access to a diverse pool of technologies, but the ability to absorb these technologies determines the extent to which they contribute to innovation outcomes like new product development and process improvements.
Based on this foundation, I propose the first hypothesis: H1: Technology absorption capacity is positively correlated with innovation performance in electric vehicle enterprises in China. This hypothesis aligns with empirical evidence from other sectors, where higher absorption capacity has been linked to increased innovation outputs. For instance, in the pharmaceutical industry, firms with strong absorption capabilities demonstrate superior innovation performance by integrating external research into their development pipelines. Similarly, in the China EV context, firms that excel in absorbing alliance technologies are likely to achieve better innovation metrics, such as higher R&D intensity and more patented innovations.
Next, I consider the role of market competition, which refers to the intensity of rivalry among firms within the electric vehicle market in China. Market competition can influence how firms leverage their absorption capacity; in highly competitive environments, firms may face pressures that alter their innovation strategies. Some studies suggest that intense competition can stimulate innovation by forcing firms to seek advantages, while others argue that it may lead to short-termism, reducing the focus on long-term absorptive activities. In the electric vehicle industry, where China EV companies operate in a crowded and fast-paced market, competition could moderate the relationship between absorption capacity and innovation performance. Thus, I hypothesize: H2: Market competition negatively moderates the relationship between technology absorption capacity and innovation performance in electric vehicle enterprises in China. This implies that as competition intensifies, the positive effect of absorption capacity on innovation may weaken, as firms divert resources to immediate survival tactics rather than sustained knowledge absorption.
Furthermore, government subsidies play a significant role in the innovation ecosystem of China’s electric vehicle sector. These subsidies, often provided to support R&D and adoption of new technologies, can act as a bridge between absorption capacity and innovation performance. By alleviating financial constraints, government subsidies enable firms to invest in advanced equipment and talent, enhancing their ability to absorb and apply external technologies. Empirical studies in the China EV industry have shown that subsidies positively impact innovation outcomes, such as increased patent filings and improved product quality. Therefore, I propose: H3: Government subsidies mediate the relationship between technology absorption capacity and innovation performance in electric vehicle enterprises in China. This mediation suggests that absorption capacity not only directly boosts innovation but also indirectly does so by increasing the likelihood of receiving government support, which in turn fuels further innovative activities.
In summary, this literature review highlights the interconnectedness of technology absorption capacity, market competition, and government subsidies in driving innovation within the electric vehicle sector in China. The subsequent sections will test these hypotheses using empirical data, providing insights into how China EV firms can optimize their alliance participation for better performance.
Research Design and Methodology
To empirically examine the relationships outlined in the hypotheses, I utilized a dataset comprising Chinese electric vehicle companies that are members of the Electric Vehicle Industry Technology Innovation Strategic Alliance. The sample period spans from 2009 to 2023, focusing on publicly listed firms in Shanghai and Shenzhen stock exchanges. Data were sourced from reliable financial databases, including the Tonghuashun and CSMAR databases, ensuring comprehensive coverage of relevant variables. After applying filters to exclude firms with special treatment statuses (e.g., ST, *ST) and adjusting for outliers by winsorizing continuous variables at the 1st and 99th percentiles, the final sample consisted of 69 electric vehicle enterprises, yielding 492 observations in an unbalanced panel format. This sample represents a significant portion of the China EV landscape, allowing for robust analysis.
The variables used in this study are defined as follows, with detailed descriptions provided in Table 1. The dependent variable, innovation performance (R&D), is measured as the ratio of R&D expenditure to total assets, reflecting the firm’s commitment to innovation activities. This metric is commonly used in literature to capture innovation input, which is less susceptible to external fluctuations compared to output measures like patents. The independent variable, technology absorption capacity (AC), is quantified as the ratio of R&D investment to operating revenue, indicating the firm’s ability to assimilate external technologies. Control variables include firm size (Size, measured as the natural logarithm of total assets), leverage ratio (Lev, total liabilities divided by total assets), dynamic capability (DC, net profit divided by total assets), and executive compensation (Pay, natural logarithm of the average salary of the top three executives). These controls account for factors that might influence innovation performance, such as resource availability and managerial incentives.
| Variable Type | Variable Name | Symbol | Description |
|---|---|---|---|
| Dependent Variable | Innovation Performance | R&D | R&D Expenditure / Total Assets |
| Independent Variable | Technology Absorption Capacity | AC | R&D Investment / Operating Revenue |
| Control Variables | Firm Size | Size | Natural Logarithm of Total Assets |
| Leverage Ratio | Lev | Total Liabilities / Total Assets | |
| Dynamic Capability | DC | Net Profit / Total Assets | |
| Executive Compensation | Pay | Natural Logarithm of Top Three Executives’ Average Salary |
To test the hypotheses, I employed fixed-effects panel regression models, which control for unobserved time-invariant characteristics of firms and time-specific effects. The baseline model is specified as follows:
$$ R\&D_{i,t} = \alpha_0 + \alpha_1 AC_{i,t} + \alpha_2 Size_{i,t} + \alpha_3 Lev_{i,t} + \alpha_4 DC_{i,t} + \alpha_5 Pay_{i,t} + \mu_i + v_t + \epsilon_{i,t} $$
where \( i \) denotes the firm, \( t \) denotes the year, \( \mu_i \) represents firm fixed effects, \( v_t \) represents year fixed effects, and \( \epsilon_{i,t} \) is the error term. This model allows me to estimate the direct effect of technology absorption capacity on innovation performance while accounting for firm-specific and temporal factors. For the moderation analysis involving market competition (MC), I extended the model to include an interaction term between AC and MC, measured as the ratio of selling expenses to operating revenue. The moderated model is:
$$ R\&D_{i,t} = \alpha_0 + \alpha_1 AC_{i,t} + \alpha_2 MC_{i,t} + \alpha_3 AC_{i,t} \times MC_{i,t} + \alpha_4 Size_{i,t} + \alpha_5 Lev_{i,t} + \alpha_6 DC_{i,t} + \alpha_7 Pay_{i,t} + \mu_i + v_t + \epsilon_{i,t} $$
For the mediation analysis, I followed the approach by Wen and Ye (2014) to test the role of government subsidies (Sub), measured as the ratio of government grants related to R&D to operating revenue. The mediation models are:
$$ Sub_{i,t} = \beta_0 + \beta_1 AC_{i,t} + \beta_2 Size_{i,t} + \beta_3 Lev_{i,t} + \beta_4 DC_{i,t} + \beta_5 Pay_{i,t} + \mu_i + v_t + \epsilon_{i,t} $$
$$ R\&D_{i,t} = \gamma_0 + \gamma_1 AC_{i,t} + \gamma_2 Sub_{i,t} + \gamma_3 Size_{i,t} + \gamma_4 Lev_{i,t} + \gamma_5 DC_{i,t} + \gamma_6 Pay_{i,t} + \mu_i + v_t + \epsilon_{i,t} $$
These models enable me to assess whether government subsidies serve as a pathway through which technology absorption capacity influences innovation performance in the China EV industry. All regression analyses were conducted using statistical software, with robust standard errors clustered at the firm level to address potential heteroskedasticity and autocorrelation.
Empirical Results and Analysis
The descriptive statistics for all variables are summarized in Table 2. The mean innovation performance (R&D) is 0.031, with a standard deviation of 0.017, indicating variation in how much firms invest in R&D relative to their assets. The minimum and maximum values range from -0.002 to 0.090, suggesting that some electric vehicle enterprises in China prioritize innovation more than others. Technology absorption capacity (AC) has a mean of 4.703 and a standard deviation of 2.274, with values spanning from -0.330 to 13.300, reflecting diverse capabilities in absorbing external technologies among China EV firms. The control variables exhibit distributions consistent with prior studies; for instance, firm size (Size) averages 23.120, and leverage (Lev) has a mean of 0.463, indicating moderate debt levels.
| Variable | Mean | Standard Deviation | Minimum | Maximum |
|---|---|---|---|---|
| Innovation Performance (R&D) | 0.031 | 0.017 | -0.002 | 0.090 |
| Technology Absorption Capacity (AC) | 4.703 | 2.274 | -0.330 | 13.300 |
| Firm Size (Size) | 23.120 | 1.551 | 20.421 | 26.961 |
| Leverage Ratio (Lev) | 0.463 | 0.158 | 0.112 | 0.762 |
| Dynamic Capability (DC) | 0.038 | 0.044 | -0.143 | 0.150 |
| Executive Compensation (Pay) | 13.678 | 0.730 | 12.267 | 15.776 |
Correlation analysis, presented in Table 3, reveals a significant positive relationship between technology absorption capacity and innovation performance (correlation coefficient of 0.727, significant at the 1% level), providing preliminary support for H1. The control variables show expected correlations; for example, firm size is positively correlated with innovation performance, while leverage has a mixed association. To check for multicollinearity, I computed variance inflation factors (VIF), as shown in Table 4. All VIF values are below 10, with the highest being 2.450 for firm size, indicating that multicollinearity is not a concern in the regression models.
| Variable | R&D | AC | Size | Lev | DC | Pay |
|---|---|---|---|---|---|---|
| R&D | 1.000 | |||||
| AC | 0.727*** | 1.000 | ||||
| Size | 0.087*** | -0.084** | 1.000 | |||
| Lev | 0.141*** | -0.034** | 0.639*** | 1.000 | ||
| DC | 0.083*** | -0.096** | 0.059*** | -0.244*** | 1.000 | |
| Pay | 0.265*** | -0.111** | 0.608*** | -0.376*** | 0.123*** | 1.000 |
Note: * p<0.10, ** p<0.05, *** p<0.01
| Variable | VIF | 1/VIF |
|---|---|---|
| AC | 1.070 | 0.938 |
| Size | 2.450 | 0.408 |
| Lev | 1.950 | 0.512 |
| DC | 1.190 | 0.841 |
| Pay | 1.680 | 0.594 |
The regression results for the baseline model are reported in Table 5. In column (1), without control variables, technology absorption capacity has a coefficient of 0.004, significant at the 1% level. After including controls in column (2), the coefficient remains positive and significant at 0.005, confirming H1. This implies that a one-unit increase in technology absorption capacity is associated with a 0.005-unit increase in innovation performance, holding other factors constant. Among the controls, firm size has a negative coefficient (-0.007, significant at 10%), suggesting that larger electric vehicle firms in China might have lower R&D intensity, possibly due to bureaucratic inefficiencies. Leverage and dynamic capability show positive effects, indicating that firms with higher debt or profitability may invest more in innovation. Executive compensation is also positively related, aligning with incentive theories. The R-squared value of 0.563 in column (2) indicates that the model explains a substantial portion of the variance in innovation performance.
| Variable | (1) R&D | (2) R&D |
|---|---|---|
| AC | 0.004*** (0.001) | 0.005*** (0.001) |
| Size | -0.007* (0.004) | |
| Lev | 0.032*** (0.009) | |
| DC | 0.032*** (0.011) | |
| Pay | 0.004*** (0.001) | |
| Constant | 0.024** (0.010) | 0.100 (0.073) |
| Firm Fixed Effects | Yes | Yes |
| Year Fixed Effects | Yes | Yes |
| Observations | 492 | 492 |
| R-squared | 0.422 | 0.563 |
Note: Robust standard errors in parentheses; * p<0.10, ** p<0.05, *** p<0.01
To ensure the robustness of these findings, I conducted additional tests. First, I replaced the measure of technology absorption capacity with the natural logarithm of R&D investment, as shown in Table 6, column (1). The coefficient remains positive and significant (0.023, p<0.01), supporting H1. Second, I randomly reduced the sample size to 419 observations in column (2), and the result holds (coefficient 0.005, p<0.01). These checks confirm that the relationship between absorption capacity and innovation performance is robust to alternative specifications and sample variations in the China EV industry.
| Variable | (1) R&D | (2) R&D |
|---|---|---|
| AC (Alternative Measure) | 0.023*** (0.002) | |
| AC (Original Measure) | 0.005*** (0.001) | |
| Size | -0.024*** (0.003) | -0.006 (0.004) |
| Lev | 0.008 (0.006) | 0.033*** (0.010) |
| DC | -0.007 (0.007) | 0.030** (0.013) |
| Pay | 0.001 (0.001) | 0.004** (0.002) |
| Constant | 0.141*** (0.041) | 0.087 (0.082) |
| Firm Fixed Effects | Yes | Yes |
| Year Fixed Effects | Yes | Yes |
| Observations | 492 | 419 |
| R-squared | 0.803 | 0.575 |
Heterogeneity Analysis
To explore variations in the relationship between technology absorption capacity and innovation performance, I conducted subgroup analyses based on ownership type and regional distribution within the China EV sector. The results, presented in Table 7, reveal significant heterogeneity. For ownership, state-owned enterprises (SOEs) show a stronger positive effect (coefficient 0.005, p<0.01) compared to non-SOEs (coefficient 0.003, p<0.01). This may be due to SOEs’ better access to resources and longer-term orientation in innovation alliances. Regionally, firms in central China exhibit the highest coefficient (0.005, p<0.01), followed by eastern and western regions (both 0.004, p<0.01). The central region’s advantage could stem from earlier establishment of innovation alliances, fostering a more conducive environment for knowledge absorption and application in the electric vehicle industry.
| Variable | (1) SOEs R&D | (2) Non-SOEs R&D | (3) Eastern R&D | (4) Central R&D | (5) Western R&D |
|---|---|---|---|---|---|
| AC | 0.005*** (0.001) | 0.003*** (0.000) | 0.004*** (0.001) | 0.005*** (0.001) | 0.004*** (0.001) |
| Size | -0.008* (0.004) | -0.008** (0.004) | -0.008* (0.004) | 0.010*** (0.001) | 0.001 (0.006) |
| Lev | 0.038*** (0.012) | 0.023** (0.010) | 0.033*** (0.009) | -0.020 (0.023) | 0.018 (0.020) |
| DC | 0.042** (0.020) | 0.020* (0.011) | 0.031*** (0.012) | -0.048 (0.059) | 0.004 (0.032) |
| Pay | 0.002 (0.003) | 0.006*** (0.002) | 0.004*** (0.002) | -0.000 (0.002) | 0.004 (0.004) |
| Constant | 0.134 (0.094) | 0.114 (0.071) | 0.120 (0.088) | -0.197*** (0.038) | -0.086 (0.091) |
| Firm Fixed Effects | Yes | Yes | Yes | Yes | Yes |
| Year Fixed Effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 231 | 261 | 411 | 40 | 41 |
| R-squared | 0.605 | 0.635 | 0.556 | 0.807 | 0.861 |
Mechanism Analysis: Moderation and Mediation Effects
To delve deeper into the underlying mechanisms, I examined the moderating role of market competition and the mediating effect of government subsidies. For moderation, I introduced an interaction term between technology absorption capacity and market competition into the regression model. The results, shown in Table 8, indicate that the interaction term has a negative coefficient (-0.010, p<0.05), supporting H2. This implies that as market competition intensifies in the China EV industry, the positive impact of absorption capacity on innovation performance diminishes. In practical terms, highly competitive environments may force electric vehicle firms to focus on immediate gains rather than long-term knowledge absorption, thereby weakening the benefits of alliance participation.
| Variable | (1) R&D | (2) R&D |
|---|---|---|
| AC | 0.005*** (0.001) | 0.005*** (0.001) |
| MC | -0.016 (0.042) | |
| AC × MC | -0.010** (0.005) | |
| Size | -0.007* (0.004) | -0.007* (0.004) |
| Lev | 0.032*** (0.009) | 0.032*** (0.009) |
| DC | 0.032*** (0.011) | 0.030*** (0.011) |
| Pay | 0.004*** (0.001) | 0.004*** (0.001) |
| Constant | 0.100 (0.073) | 0.100 (0.071) |
| Firm Fixed Effects | Yes | Yes |
| Year Fixed Effects | Yes | Yes |
| Observations | 492 | 492 |
| R-squared | 0.563 | 0.577 |
For mediation, I applied a three-step approach to test the role of government subsidies. As reported in Table 9, technology absorption capacity positively affects government subsidies (coefficient 0.001, p<0.10 in column 2), and when both absorption capacity and subsidies are included in the innovation performance model (column 3), subsidies show a negative but significant coefficient (-0.108, p<0.05), while absorption capacity remains positive (0.005, p<0.01). This partial mediation supports H3, indicating that absorption capacity enhances innovation performance partly by increasing the likelihood of receiving government subsidies. However, the negative coefficient for subsidies in the full model suggests that in some cases, subsidies might not directly translate to better innovation if not coupled with effective absorption, highlighting the complexity of policy impacts in the China EV context.
| Variable | (1) R&D | (2) Sub | (3) R&D |
|---|---|---|---|
| AC | 0.005*** (0.001) | 0.001* (0.0009) | 0.005*** (0.001) |
| Sub | -0.108** (0.051) | ||
| Size | -0.007* (0.004) | 0.003 (0.003) | -0.006* (0.004) |
| Lev | 0.032*** (0.009) | -0.020 (0.014) | 0.030*** (0.008) |
| DC | 0.032*** (0.011) | 0.011 (0.016) | 0.033*** (0.011) |
| Pay | 0.004*** (0.001) | 0.0004 (0.002) | 0.004*** (0.001) |
| Constant | 0.100 (0.073) | -0.036 (0.076) | 0.096 (0.072) |
| Firm Fixed Effects | Yes | Yes | Yes |
| Year Fixed Effects | Yes | Yes | Yes |
| Observations | 492 | 492 | 492 |
| R-squared | 0.563 | 0.375 | 0.575 |
Conclusion and Recommendations
In conclusion, this study demonstrates that technology absorption capacity is a key driver of innovation performance in China’s electric vehicle industry, particularly within innovation alliances. The empirical evidence supports a positive correlation, with heterogeneity across ownership types and regions, where state-owned enterprises and firms in central China benefit more significantly. Additionally, market competition acts as a negative moderator, while government subsidies serve as a partial mediator in this relationship. These findings offer valuable insights for electric vehicle companies in China, often referred to as China EV firms, as they navigate the complexities of alliance participation and innovation management.
Based on these results, I recommend that electric vehicle enterprises prioritize enhancing their technology absorption capabilities by investing in R&D and fostering open innovation practices. For alliance organizations, it is crucial to curate partnerships with firms that have strong resource bases, facilitating knowledge sharing and technological integration. Given the heterogeneity findings, policymakers and alliance managers should leverage the exemplary role of state-owned enterprises and learn from the central region’s experiences to promote balanced regional development in the China EV sector. Furthermore, firms must carefully manage market competition risks by monitoring industry dynamics and avoiding short-term pressures that could undermine long-term absorption activities. Government subsidies should be strategically utilized as a catalyst, but firms need to ensure that these funds are coupled with robust absorption mechanisms to maximize innovation outcomes.
Overall, this research underscores the importance of technology absorption capacity in the electric vehicle industry of China, providing a framework for firms to achieve sustainable growth through strategic alliances. Future studies could explore additional moderators or extend the analysis to other emerging technologies within the China EV ecosystem, further enriching our understanding of innovation dynamics.