2.1 University R&D spillover and the performance of agribusiness firms

It is generally believed that spillover effects are caused by the non-competitive and non-exclusive nature of R&D knowledge, and they are an important method by which enterprises acquire knowledge from other organizations [11, 24]. As one of the primary providers of public R&D, universities play a significant role for enterprises in obtaining external knowledge [13]. There are diverse pathways for university agricultural R&D to influence agribusiness firms, which is essential for driving innovation and securing a competitive edge. Collaborative research initiatives and partnerships often serve as a primary channel for this knowledge transfer, with firms and universities engaging in joint research projects that yield mutual benefits [25, 26]. The transition of university graduates and researchers into the industry is another critical vector, infusing fresh expertise and state-of-the-art practices into the agribusiness sector [27]. Informal networks and industry events, including conferences and workshops, further facilitate the exchange of ideas and the latest findings in agricultural R&D [28]. Access to new research outputs is also gained through academic publications and patents, which agribusiness firms can utilize to stay abreast of technological advancements [29]. Additionally, universities often extend consultancy and advisory services, translating complex research findings into practical solutions for business challenges [30].

Qin and Du [31] and Fukugawa [32] found that university R&D spillovers impacted enterprises in a specific area. Furthermore, Zhang, et al. [33] examined the impact of university R&D on Chinese manufacturing firms to validate the relationship between university spillovers and corporate performance. Moreover, Woodward et al. [34] demonstrated a link between university spillover and corporate performance using R&D expenditure at universities as the indicator. Ponds et al. [35] confirmed the impact of university R&D spillover on enterprises from the perspective of university-industry collaboration. More importantly, Anousheh et al. [10] show that university R&D spillovers contribute to the changes in agricultural output. Accordingly, it can be concluded that university R&D spillovers are influential to the performance of agribusiness firms to a certain degree. However, the assimilation of this knowledge is not without its barriers; organizational inertia can significantly dampen a firm’s ability to adapt to and adopt new innovations, necessitating deliberate strategies to mitigate these effects and fully capitalize on the potential of university R&D for sustainable growth and innovation.

2.2 Agricultural R&D from university and short-term performance of agribusiness firms

While the existence and impacts of university R&D spillovers on various industries have been well-documented, the specific dynamics between these spillovers and agribusiness firms remain an area ripe for exploration. The intricate interplay between external knowledge and internal organizational structures is pivotal in understanding this relationship.

Liao et al. [20] and Lin et al. [21] delve into the concept of organizational inertia, which is rooted in entrenched problem-solving methodologies derived from a firm’s existing knowledge base. Organizational inertia refers to a firm’s resistance to change due to established routines, processes, and structures. When confronted with new agricultural R&D spillovers, agribusiness firms with high inertia might be slow to adapt or even reject these innovations, leading to short-term performance declines. This inertia, in essence, acts as a barrier, making enterprises less receptive to external R&D spillovers. Delving deeper into the psychological dimensions, Godkin and Allcorn [36] highlight that such inertia can foster a defensive mindset within organizations. This defensiveness can manifest in various dysfunctions at both individual and group levels, ultimately casting a shadow on overall corporate performance. Without proactive organizational interventions, this inertia-driven resistance can persist, leading to suboptimal performance outcomes in the short run.

Assink [37] provides further nuance by illustrating how a firm’s historical experiences and accumulated knowledge can, paradoxically, become impediments, especially when it comes to innovation performance. Firms with a history of success in certain practices might be reluctant to deviate from established paths. The introduction of new R&D findings can challenge these established practices, causing firms to hesitate or delay their adoption. Vecchiat [37] echoes this sentiment, suggesting that a firm’s legacy knowledge can sometimes muddle its adaptability to emerging trends and novel challenges. Lin et al. [21] introduce an intriguing dimension to this discussion by positing that the closer the alignment between a firm’s internal knowledge and incoming external knowledge, the more pronounced the adverse effects of inertia become.

Given the inherent parallels between the knowledge generated from university agricultural research and the foundational knowledge within agribusiness firms, it’s plausible to hypothesize a nuanced relationship between the two. From the perspective of organizational inertia, the negative short-term relationship between agricultural R&D spillovers from universities and agribusiness firm performance can be attributed to the inherent resistance to change within established firms. Therefore, in the short term, the influx of agricultural R&D spillovers from universities may clash with the entrenched knowledge structures within agribusiness firms, leading to a potential negative impact on their performance. The first hypothesis is put forth as follows:

1. Hypothesis 1: The university’s agricultural R&D spillovers are negatively related to the performance of agribusiness firms in the short term.

2.3 Moderating effect of university non-agricultural R&D spillovers

Agribusiness firms are influenced by a myriad of knowledge sources, including both agricultural and non-agricultural R&D spillovers emanating from universities. While the knowledge derived from agricultural R&D is inherently familiar to these firms, it often leads to a phenomenon known as organizational inertia. This inertia, rooted in the familiarity and reliance on existing knowledge, can impede innovation, making the direct benefits of agricultural R&D spillovers less pronounced in terms of firm performance.

In juxtaposition, non-agricultural R&D spillovers introduce a diverse and novel knowledge spectrum. Given that this knowledge is not entrenched in the firms’ existing knowledge base, it bypasses the barriers of organizational inertia. Non-agricultural R&D often brings fresh perspectives, methodologies, and technologies that can be applied innovatively in the agricultural sector. For instance, advancements in IT or materials science can lead to the development of better farming equipment, precision agriculture tools, or improved storage solutions [38]. Drawing from the insights of Yang and Maxwell [39], there’s an inherent value in diverse knowledge, and such valuable knowledge tends to disseminate more effectively within an organization. This is further echoed by Gupta and Govindarajan [22], who emphasized the fluidity of high-value knowledge within organizational structures. A broader R&D base, encompassing both agricultural and non-agricultural research, provides agribusiness firms with a diversified knowledge pool. This diversification can lead to more robust solutions, combining the best of both worlds. Building on this, Monteiro et al. [23] and Lai et al. [40] posited that the effective flow and assimilation of such valuable knowledge can catalyze innovation. This innovation, in turn, is bolstered by the acquisition of diverse knowledge, enhancing overall firm performance.

In essence, while agricultural R&D provides domain-specific knowledge crucial for agribusiness firms, non-agricultural R&D spillovers offer complementary insights, technologies, and practices that can enhance and positively moderate the impact of agricultural R&D on firm performance. The synergy between the two can lead to holistic solutions that drive superior outcomes. Consequently, it can be inferred that non-agricultural R&D spillovers, by virtue of their novelty and value, can amplify the positive effects of agricultural R&D spillovers on agribusiness firms. This amplification arises from the stimulation of high-value knowledge flows, fostering an environment conducive to innovation and knowledge acquisition, which ultimately drives superior enterprise performance.

1. Hypothesis 2: University non-agricultural R&D spillovers positively moderate the relationship between agricultural R&D spillovers and the performance of agribusiness firms.

2.4 The long-term effect of university agricultural R&D

Agribusiness firms, when first exposed to university agricultural R&D spillovers, often exhibit organizational inertia. This inertia can lead to an initial resistance to the integration of new knowledge, resulting in short-term performance setbacks. However, it’s essential to understand that this inertia is not a static phenomenon. Over time, various enterprise activities can influence and reduce the effects of this inertia.

Gnyawali and Park [41] highlighted the role of co-opetition, a strategy where firms both collaborate and compete, in helping companies navigate and overcome the barriers posed by organizational inertia. In a similar vein, Huang et al. [14] emphasized that broadening the scope of the innovation process can counteract the detrimental effects of inertia, fostering a more conducive environment for organizational growth and adaptation. Overtime, the innovations from R&D spillovers can lead to increased yields, better quality products, and more efficient farming practices.

Further insights from Easterby-Smith and Lyles [42] and Huang et al. [17] introduced the concept of organizational forgetting. This process, where firms intentionally or unintentionally let go of certain knowledge or practices, can be instrumental in diminishing the constraints of organizational inertia. However, it’s pivotal to note that such transformative enterprise activities, including organizational forgetting, exploratory learning, and expansive innovation, are not instantaneous. As Huang et al. [17] elucidated, these processes require a significant duration to manifest fully, implying that their inertia-mitigating effects are inherently long-term. A recent study by Su et al. [43] reveals a U-shaped relationship between agribusiness diversification and technological innovation efficiency in China. The findings emphasize the importance for agribusiness managers to adopt a long-term perspective on diversification. This approach counters the risk of "managerial myopia," which might deter them from investing in R&D projects that have uncertain returns in the long run.

In essence, while the initial phase of integrating university agricultural R&D spillovers can present challenges and costs, the long-term benefits in terms of innovation, efficiency, market positioning, and sustainability can significantly outweigh these short-term hurdles, leading to enhanced performance and growth for agribusiness firms. Given the above dynamics, it becomes evident that while the immediate aftermath of university agricultural R&D spillovers might be met with resistance due to inertia, the long-term landscape is markedly different. The enduring positive contributions of these spillovers to the agricultural sector, combined with the time-dependent reduction in organizational inertia within agribusiness firms, suggest a more favorable long-term impact of university agricultural R&D spillovers compared to the short-term.

1. Hypothesis 3: The university agricultural R&D spillovers have positive effects on the performance of agribusiness firms in the long term.

2.5 The effect of geographical proximity on university agricultural R&D spillovers

R&D spillover exhibits characteristics of geographic proximity due to the nature of knowledge flow, which is dependent on individual behavior [35, 44]. Consequently, the effect of R&D spillover may diminish as distance increases, subsequently reducing its impact on company performance. Notably, this geographical proximity is also present in university R&D spillovers. Proximity to universities often facilitates easier and more frequent interactions between researchers and agribusiness professionals. As distance increases, the ease of direct communication and collaboration can decrease, leading to potential barriers in effective knowledge transfer.

Lagendijk and Lorentzen [45] study indicates that shorter geographic distances foster enhanced interaction between two organizations, thereby increasing their knowledge-sharing cooperation. Similarly, Antonelli [46] assert that close geographical proximity facilitates face-to-face communication between organizations, potentially promoting enterprise innovation by disseminating tacit knowledge. In summary, while distance is just one factor, it plays a significant role in determining the extent and effectiveness of agricultural R&D spillovers from universities to agribusiness firms. Proximity facilitates better communication, collaboration, resource sharing, and trust-building, all of which are crucial for maximizing the benefits of R&D spillovers. Based on these findings, the following hypothesis is proposed in this study:

1. Hypothesis 4: The effect of university agricultural R&D spillovers is negatively related to the increase in distance between agribusiness firms and universities.

3.1 Data collection

This study utilized a unique panel dataset spanning from 2009 to 2019. Firm-level data were sourced from the China Stock Market & Accounting Research (CSMAR) database, while university R&D spillover data were collected using Google Earth. For subsequent analysis, Chinese universities were classified based on their areas of expertise. To evaluate the effects of agricultural R&D spillovers, agricultural universities specializing in the same fields as agribusiness firms were selected. Non-agricultural R&D spillovers were measured by considering science and technology universities, which supply heterogeneous knowledge to agribusiness firms. To mitigate potential bias arising from the 2008 financial crisis and the COVID-19 pandemic, a panel data analysis was conducted on China’s listed agribusiness firms from 2009 to 2019 to examine the impact of university R&D spillovers. Given the extended period required for enterprises to absorb R&D spillovers, all university R&D spillover data in this study were lagged by one year in the regression analysis.

The previous study posits that agribusiness is inherently linked to value chains, which not only end with the product’s arrival at the consumer’s table but also encompass all the value-adding operations required along the way [47, 48]. Therefore, our definition of agribusiness firms includes entities involved in the production, processing, distribution, and retailing of agricultural products and services [43, 49]. The panel dataset for analysis was constructed as follows: (1) Firm years for which the registered address of agribusiness firms could not be obtained were excluded. (2) Firm years with transaction statuses of particular transfer (PT) were removed to avoid potential errors caused by firms’ improvement in listing status through increased discretionary accrual. (3) Firm years with missing data for any variables planned for analysis during the study period were eliminated. Ultimately, the aforementioned criteria yielded a total of 1,230 firm-year observations for 179 agribusiness firms across the 2009–2019 timeframe. All relevant data are included in the Supporting Information files.

3.2 Variables

3.2.1 Independent variables.

In this study, university agricultural R&D spillovers were treated as an independent variable. Given the variety of channels through which university agricultural R&D spillovers occur, quantifying their effects is challenging. Consequently, following the approach of Hilary and Hui [50], this study measured university agricultural R&D spillovers by selecting the number of agricultural universities within a certain radius of agribusiness firms’ registered addresses. Considering the characteristics of university R&D spillovers and the distribution of Chinese universities, a distance of 200 kilometers was adopted as the appropriate radius [51, 52]. The following methodology was employed to evaluate knowledge spillover effects: First, a list of agricultural enterprises was compiled by adopting the classification criteria set forth by the China Securities Regulatory Commission in 2012 and utilizing the CSMAR database. Subsequently, the registered addresses of these enterprises were identified, and their geographic coordinates (longitude and latitude) were ascertained using Google Earth. Second, a comprehensive list of agricultural universities was curated, drawing upon the official list disseminated by the Ministry of Education of the People’s Republic of China and the Airuishen alumni network. The geographic coordinates of these universities were also determined via Google Earth.

Third, to compute the distance between universities and listed agricultural enterprises from 2009 to 2019, we employed the methodology outlined in Du et al. [51], using the geographic coordinates as follows: (a) The university’s longitude and latitude were represented as longitudeU and latitudeU, respectively, while those of an agribusiness firm were designated as longitudeF and latitudeF. Utilizing these coordinates, the central angle (θ) was determined through Eq (1). (b) Subsequently, the arc length per radian was calculated using Eq (2). (c) The distance between each agricultural university and the registered address of each agribusiness firm was ascertained by applying Eq (3), as this distance corresponds to the length of a minor arc across the Earth’s surface. Ultimately, the variable UARS (university agricultural R&D spillovers) was defined based on the computed number of agricultural universities located within a 200-kilometer radius of the companies’ registered addresses.

(1)(2)(3)

3.2.4. Control variables.

Leverage (LEV): Leverage refers to the employment of borrowed funds to amplify return on equity. The influence of leverage on corporate performance has been a subject of debate. While Öhman [53] posits that high leverage can negatively impact corporate performance by decreasing R&D investment, increasing leverage can enhance innovation and performance efficiency for companies experiencing low growth rates. In this study, the leverage of listed agricultural companies is measured using the ratio of total liabilities to total assets.

Equity Nature (EN): The effect of equity nature on corporate performance remains inconclusive in previous research. Choi et al. [54] suggest that state-owned enterprises with close government ties can more easily secure financial support, thus bolstering R&D investment and performance. Conversely, Lin [55] contends that non-state-owned enterprises may outperform state-owned counterparts when utilizing long-term bank loans, although this has not been definitively established for state-owned enterprises. Consequently, equity nature is employed as a control variable to analyze the impact of university R&D spillovers on agribusiness firms. The variable value is set at 1 for state-owned enterprises and 0 for others.

Ownership Concentration (OC): Ownership concentration refers to the proportion of shares held by a company’s major shareholder. A higher ownership concentration may enhance corporate governance, promoting innovation efficiency and improving corporate performance [56]. In this study, ownership concentration is calculated as the ratio of shares held by the largest shareholder to the total number of company shares.

GDP Level (GDP): GDP level serves as an indicator of a region’s overall economic performance. Enterprises situated in developed regions tend to exhibit superior performance compared to those in less developed areas. Drawing on prior research, this study utilizes the natural logarithm of the GDP of the province in which a company operates to evaluate the impact of regional growth on performance. The definitions and sources for all variables involved in the study are provided in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Definition of variables and data source.

https://doi.org/10.1371/journal.pone.0296007.t001

3.3 Models

In this investigation, the influence of university R&D spillovers on the performance of agribusiness firms was examined through the employment of fixed-effects models. The foundational model utilized in this study is delineated as follows. The Hausman test outcome leads to the rejection of the null hypothesis, signifying that fixed-effects regression constitutes the favored methodology. For comprehensive review and reference purposes, the baseline results of the random effect regression analysis have also been detailed and included in a table in the Appendix in supporting information files. In this study, all statistical analyses were conducted using Stata 17.

(4)

(5)

In this context, UARS i, t-1, and UNARS i, t-1 denote the university agricultural R&D spillovers and non-agricultural spillovers, respectively, for firm i in the preceding year (t-1). Performance i, t encompasses variables associated with the performance of agribusiness firm i during year t. Concurrently, Z i, t represents the control variables influencing the performance of agribusiness firms, while ε i, t corresponds to the random disturbance terms.

4.1 Descriptive analysis and correlation analysis

Table 2 presents descriptive statistics for the primary variables examined in this study, utilizing panel data from 2009 to 2019. The average and median values of performance (ROA) are both 0.05, suggesting a normal distribution for ROA. The mean and standard deviation for the number of agricultural universities situated within a 200-kilometer radius of the listed agribusiness firms are 2.07 and 1.62, respectively. However, UNARS exhibits significant variation, with a standard deviation of 11.63. In relation to the other control variables, the mean leverage ratio stands at 0.38, which is deemed acceptable. The average value of EN is 0.37, signifying that 37% of agricultural enterprises are state-owned entities.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Descriptive statistics.

https://doi.org/10.1371/journal.pone.0296007.t002

Table 3 displays the correlation coefficients for the variables under study. In line with expectations, the performance of agribusiness firms exhibits a negative and significant association with university agricultural R&D spillover (-0.058), while bearing an insignificant relationship with university non-agricultural R&D spillover (0.043). Furthermore, the performance of agribusiness firms is positively correlated with the shareholding ratio of the largest shareholder (OC) and GDP, and negatively associated with the state-owned enterprise dummy variable (EN) and the leverage ratio (LEV). To further evaluate the possibility of multicollinearity among the test and control variables, variance inflation factor (VIF) tests were conducted on the regression models. Table 4 demonstrates that no evidence of multicollinearity was found, with VIF values ranging from 1.03 to 3.23.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Correlation matrix.

https://doi.org/10.1371/journal.pone.0296007.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Variance inflation factor.

https://doi.org/10.1371/journal.pone.0296007.t004

4.2 Empirical analysis

4.2.1. The short-term effect of university agricultural R&D spillovers.

Prior research has established that university agricultural R&D spillovers are beneficial to the agricultural industry. However, this study unveils a contrasting outcome in the short term from the vantage point of agribusiness firms. Table 5 delves into the impact of university agricultural R&D spillovers on the performance of listed agribusiness firms. Column 1 presents the simple regression between university agricultural R&D spillovers (UARS) and agribusiness firms’ performance. In order to ensure robustness, we conducted a further analysis using the two-step dynamic system-GMM estimator, as recommended by Arellano and Bover (1995) [57] Arellano & Bover (1995) and Blundell and Bond [58]. The results of this analysis are presented in Column 2 of Table 5. The coefficient of Column 1 and Column 2, which is statistically significant at both the 1% and 10% levels, suggests a negative impact from university agricultural R&D spillovers (UARS)." This outcome supports Hypothesis 1, positing that the agricultural R&D efforts emanating from universities do not enhance the performance of agribusiness firms in the short term. In another word. The university’s agricultural R&D spillovers are negatively related to the performance of agribusiness firms in the short term. These findings align with Lin et al. [21], who discovered that organizational inertia adversely affects an organization’s performance.

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Regression results of the short-term effect of university R&D spillover.

https://doi.org/10.1371/journal.pone.0296007.t005

Tables 6 and 7 explores the role of university non-agricultural R&D spillovers in the relationship between agribusiness firms’ performance and agricultural R&D spillovers, as shown in column 1. By incorporating an interaction term into the model, the impact of university non-agricultural R&D spillover is examined. With a coefficient of 0.003 for the interaction term, university non-agricultural R&D spillovers (UNARS) demonstrate a positive moderating effect between agricultural R&D spillovers (UARS) and agribusiness firms’ performance. In accordance with earlier studies, the results reveal that valuable knowledge, specifically non-agricultural knowledge in this instance, fosters the performance of other types of knowledge, such as agricultural knowledge. More precisely, without the organizational inertia engendered by existing knowledge, university non-agricultural R&D spillovers for agribusiness firms can stimulate high-value knowledge within organizations, thereby increasing internal knowledge flow and enabling agricultural knowledge to yield better performance. This finding also supports Hypothesis 2, which asserts that non-agricultural R&D spillovers from universities facilitate enhanced performance of agribusiness firms by influencing agricultural R&D spillovers.

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Regression results of the short-term effect of university R&D spillover.

https://doi.org/10.1371/journal.pone.0296007.t006

Download:

• PNG
  larger image
• TIFF
  original image

Table 7. Regression results of two-period lagged value of university R&D spillover.

https://doi.org/10.1371/journal.pone.0296007.t007

In order to assess the robustness of our findings, we estimated the regression models using the density of agricultural universities and the density of science and technology universities as proxies for university agricultural R&D spillovers and non-agricultural R&D spillovers, rather than utilizing the number of agricultural universities and the number of science and technology universities. The newly derived variables UARS (D) and UNARS (D) were computed by dividing the number of a specific type of university located within 200 kilometers of the agribusiness firms by the total number of such universities in China. Dividing by the total number of such universities in China was an attempt to normalize the data, ensuring that the influence of universities is considered relative to their overall presence in the country. This approach provides a standardized metric that allows for comparative analysis across different regions and agribusiness firms. The results presented in Tables 6 and 7 reveal that the estimated coefficients for all key independent variables align with those reported in columns 1.

Moreover, we employed the two-period lagged value of the key independent variables as measures of university agricultural R&D spillovers and non-agricultural R&D spillovers to perform a robustness check. Following Ullah et al. [59] study, lagged values were utilized to address the endogeneity issue stemming from two-way causality. As illustrated in Table 7, the coefficients on UARSt-2 are negative and statistically significant in columns 4 and 5. Additionally, the coefficient on UNARSt-2 is insignificant, while the interaction term’s coefficient is positive and statistically significant. These results suggest that university agricultural R&D spillovers are negatively associated with agribusiness firms’ performance in the short term, and university non-agricultural R&D spillovers positively moderate the relationship between agricultural R&D spillovers and agribusiness firms’ performance. This corroborates our findings.

4.2.2 The long-term effect of university agricultural R&D spillovers.

Previous studies have demonstrated that factors such as innovation and organizational forgetting can weaken organizational inertia [14, 17]. Consequently, the adverse impact of organizational inertia on university agricultural R&D spillovers may be affected by other factors over time. However, these factors typically require a relatively extended period to exert a substantial influence on enterprises. As a result, the long-term effects of university agricultural R&D spillovers are likely to differ from their short-term counterparts.

Utilizing three-period lagged data based on our sample size, we investigate the long-term effects of university R&D spillovers on the performance of agribusiness firms. While we acknowledge that the general innovation process in some industries may take more than ten years, the agribusiness sector operates within a unique context. Given the nature of agricultural cycles, innovations and their subsequent spillovers can manifest and impact firm performance within shorter time frames. Table 8 reveals that university agricultural R&D spillovers (UARSt-3) exert a statistically significant long-term effect on agribusiness firms’ performance, displaying a positive sign, which contrasts with the short-term effect. Notably, concerning non-agricultural R&D spillovers, the coefficient of UNARSt-3 becomes significant, reflecting a reduction in the disparity between non-agricultural and agricultural R&D spillovers as organizational inertia diminishes over time. Additionally, the coefficient of the interaction term is statistically significant but negative at a 1% level, implying that university non-agricultural R&D spillovers continue to moderate; however, their influence on the relationship between agricultural R&D spillovers and agribusiness firms’ performance becomes negative over time. The results presented in Table 8 support Hypothesis 3 by demonstrating that university agricultural R&D spillovers have a long-term positive impact on agribusiness firms’ performance.

Download:

• PNG
  larger image
• TIFF
  original image

Table 8. Regression results of the long-term effect of university agricultural R&D spillover.

https://doi.org/10.1371/journal.pone.0296007.t008

4.2.3 The impact of differences in geographical proximity.

The empirical analysis conducted above highlights the negative impact of university agricultural R&D spillovers on agribusiness firms. Furthermore, other factors, such as distance, may influence the effectiveness of R&D spillovers. According to Ponds et al. [35] and Qiu et al. [52], university R&D spillovers tend to weaken as distance increases.

To examine the significance of distance in the relationship between university agricultural R&D spillovers and agricultural enterprise performance, we introduce a new independent variable, UARS300, which represents the number of agricultural universities within 300 kilometers of the agribusiness firms.

As illustrated in Table 9, when using the number of universities within 300 kilometers as the independent variable, the impact on the performance of agribusiness firms is diminished compared to the regression result with the number of universities within 200 kilometers as the independent variable. This empirical evidence suggests that distance attenuates the spillover effect of agricultural R&D from universities. Additionally, as distance increases, non-agricultural R&D spillovers from universities do not seem to moderate the impact of agricultural R&D spillovers on agribusiness firms’ performance. Consequently, both the agricultural R&D spillover effect on agribusiness firms’ performance and the moderating effect of non-agricultural R&D spillovers on the relationship between agricultural R&D spillovers and agribusiness firms’ performance decline as distance expands.

Download:

• PNG
  larger image
• TIFF
  original image

Table 9. Regression results of different distance.

https://doi.org/10.1371/journal.pone.0296007.t009

5.1 Theoretical and practical implications

This study bridges theoretical exploration and practical application, revealing how university agricultural R&D spillovers shape agribusiness firm performance in China. It uncovers that organizational inertia is a significant theoretical construct that hampers short-term performance due to resistance to external knowledge [15]. Firms often adhere to established procedures and resist incorporating external knowledge derived from university R&D activities [21, 60]. Practically, this suggests a need for policy makers to encourage firms to actively disrupt established norms and foster innovation, enhancing their absorptive capacity and responsiveness to R&D spillovers. For decision makers, the long-term benefits as inertia diminishes, highlighting the positive effects of university R&D, indicating that persistence in innovation strategies eventually pays off [61, 62]. This long-term gain underscores the need for sustained investment in innovation and the cultivation of core competencies that allow firms to adapt to and capitalize on evolving business conditions, particularly in the post-COVID era.

Agriculture is a crucial component of societal sustainability in China, leading to substantial public investment in agricultural R&D to bolster the industry’s performance [63]. However, the challenge that agribusiness firms face in maximizing the output from agricultural R&D investments, as these often yield disproportionate achievements, suggests a gap that policy makers need to address [64, 65]. Therefore, the study emphasizes that non-agricultural R&D spillovers serve as a catalyst for innovation in the short term but presenting challenges in the long term. This duality suggests that agribusiness firms should adopt a strategic approach to integrating diverse knowledge sources, potentially by hiring talent from varied sectors and engaging in cross-disciplinary collaborations to maintain a balance between agricultural and non-agricultural knowledge inputs. Lastly, the practicality of these findings lies in the recommendation for agribusiness firms to forge stronger alliances with universities. Such partnerships can facilitate more effective knowledge management and ensure that firms are positioned to make the most of public R&D resources.

5.2 Study limitations, recommendations and future research

Future research could delve deeper into the mechanisms underlying the impact of university agricultural R&D spillovers on agribusiness firms’ performance. Additionally, researchers and scholars might reveal external factors that influence the relationship between agricultural R&D spillovers and the performance of agribusiness firms, thus providing a more comprehensive understanding of the interplay between these entities. While future research can provide valuable insights into the dynamics between agricultural R&D spillovers and firm performance, it is important to acknowledge that such studies may encounter several limitations. These may include difficulties in quantifying the intangible aspects of knowledge spillovers, challenges in isolating the direct impacts of R&D amid a multitude of interacting factors, and potential limitations in data accessibility, particularly proprietary or competitive information that firms may be unwilling to share.

Additionally, the ever-changing nature of agricultural technologies and fluctuating market conditions can introduce significant variability, making it hard to establish long-term causal relationships. Another limitation is the generalizability of findings, as studies may be context-specific and influenced by regional policies, market structures, and the idiosyncrasies of individual agribusiness ecosystems.

Finally, the rapid advancement and adoption of new agricultural technologies can often surpass the pace of academic research, highlighting the need for continuous and timely updates to studies in this domain. Recognizing this dynamic, it is imperative for future research to not only acknowledge these constraints but also to innovate methodological approaches that remain relevant amid such fast-paced changes. To thoroughly explore the causal dynamics between university R&D spillovers and agribusiness firm performance, future studies could benefit from adopting a diverse array of research designs. Longitudinal studies could provide insights into the temporal aspects of these relationships, case studies could offer in-depth analyses of specific contexts, and experimental designs could potentially ascertain causality with greater precision. Embracing such methodological diversity will be crucial in dissecting the intricate causative threads that weave through the fabric of university-industry collaborations in the agricultural sector.