Innovation Networks and Knowledge Diffusion Across Industries: An Empirical Study from an Emerging Economy

Abstract

This study investigates the impact of innovation networks on knowledge diffusion, aligning with the Knowledge Spillover Theory of Entrepreneurship (KSTE). It explores how these networks promote sustainability-oriented innovation and enhance corporate performance across industries, focusing on the Ghanaian context. A theoretical framework was developed using a two-dimensional fixed effects model and instrumental variable estimations. The analysis utilized longitudinal data from the World Intellectual Property Organization (WIPO) via the PENTSCOPE database from 2000 to 2023. The study reveals that robust innovation networks significantly enhance knowledge diffusion. Absorptive capacity plays a crucial mediating role, enabling firms to assimilate and apply external knowledge effectively. Additionally, organizational culture is a critical moderator, with adaptive and collaborative cultures fostering more efficient knowledge integration. The results highlight the pivotal role of innovation networks in transforming technological advancements into measurable performance outcomes, driving sustainable innovation and competitive advantage. This research extends the KSTE framework by integrating the network theory to examine how firms exploit external knowledge. It provides empirical evidence from the Ghanaian context, offering novel insights into how innovation networks can foster knowledge diffusion and sustainable development. This study’s findings contribute to theoretical discourse and practical applications, emphasizing the importance of strategic engagement in innovation networks and adaptive organizational cultures for long-term success. This research provides insights applicable to other emerging economies, highlighting how robust innovation networks can drive knowledge diffusion and sustainable development beyond the Ghanaian context.

1. Introduction

In rapidly evolving industries where competition fuels economic growth [1,2,3], innovation networks play a critical role in facilitating knowledge diffusion [4]. In his study Networks in the Innovation Process, explores the pivotal roles of knowledge networks and innovation ecosystems, emphasizing how inter-organizational relationships shape innovation dynamics [5]. He highlights the significance of firm collaboration, knowledge spillovers, and the geographic and social dimensions of innovation networks. These networks, consisting of knowledge-driven firms, support joint research and development (R&D) efforts, leading to innovative solutions [6]. As a result, innovation networks and knowledge diffusion significantly benefit industries [7]. Firms that effectively utilize internal and external knowledge sources excel in fostering creativity, modernizing management practices, and sustaining competitiveness.

While previous research has highlighted the significance of firm capabilities within innovation networks [5] and their role in knowledge diffusion [8], considerable gaps remain in understanding the complex dynamics of these networks and their impact on knowledge-driven firms. Much of the literature focuses on firms’ absorptive capacity to identify, integrate, and utilize external knowledge as a critical factor for competitiveness and innovation [9,10]. This study challenges the prevailing notion by demonstrating that absorptive capacity alone is insufficient; organizational culture plays a pivotal role in moderating the relationship between innovation networks and knowledge diffusion, providing a more nuanced perspective than previously suggested. However, limited research exists on how absorptive capacity, in conjunction with innovation networks, influences the knowledge diffusion processes of firms [11,12]. Furthermore, the impact of organizational culture and the role of diverse stakeholders within innovation networks on these dynamics remain underexplored.

In Ghana, specifically within industries, the literature on innovation networks needs to be more extensive, creating a significant gap in understanding their adoption and impact. The Knowledge Spillover Theory of Entrepreneurship (KSTE) offers a valuable theoretical perspective, suggesting that knowledge generated within firms can spill over to others, fostering innovation and economic growth, particularly in knowledge-intensive sectors [13]. Startups, in particular, are often better positioned to leverage these spillovers than established firms, which may prioritize core operations over exploiting external knowledge [13,14]. The research exploring these dynamics could provide valuable insights for firms, government agencies, and academic institutions, helping to optimize innovation networks for enhanced competitive advantage and sustained innovation.

This research contributes empirically by investigating how knowledge diffusion enhances innovation networks, addressing gaps in the literature on the impact of these networks and absorptive capacity on knowledge processes. Unlike previous studies that focus primarily on absorptive capacity as an isolated factor [9,10], this study demonstrates the interactive role of organizational culture and absorptive capacity within innovation networks, thereby expanding the theoretical understanding of knowledge diffusion dynamics. It emphasizes that effective knowledge management is crucial for long-term success [15,16]. The insights from this study can inform public policies aimed at fostering innovation ecosystems and guide industrial practices by highlighting the importance of collaborative networks and absorptive capacity in driving economic growth and competitiveness. By exploring the Knowledge Spillover Theory of Entrepreneurship (KSTE), this study develops a conceptual model linking innovation networks to knowledge diffusion, highlighting the amplifying role of absorptive capacity. Using data from the Ghana Statistical Service (GSS) and PENTSCOPE, the analysis examines how organizational culture moderates these relationships.

This study’s significance lies in its potential to inform managerial decision-making, strategic planning, and policymaking frameworks across various industries. By illustrating how innovation networks, knowledge diffusion, organizational culture, and absorptive capacity interact to enhance organizational effectiveness, this study contributes to both theoretical advancements and practical applications in business management and innovation. Additionally, the findings have broader implications for the evolving technological innovation landscape and sustainability in the global economy.

This study’s theoretical framework is based on the Knowledge Spillover Theory of Entrepreneurship (KSTE), which suggests that knowledge generated in one context can benefit entrepreneurs and firms, promoting innovation and new business creation [13]. Organizational culture and absorptive capacity are critical enablers that enhance the effectiveness of innovation networks in facilitating knowledge diffusion. This research also incorporates the network theory, emphasizing the importance of relationships among organizations for knowledge exchange and innovation diffusion [2].

By integrating organizational culture and absorptive capacity with innovation networks, this study presents a holistic approach to enhancing innovation processes, fostering collaboration, and achieving sustainable competitive advantage. As industries navigate the challenges of innovation networks and knowledge diffusion, the insights gained provide actionable strategies to address complex issues and capitalize on emerging opportunities.

The primary aim of this research is to investigate the effects of innovation networks on knowledge diffusion capabilities, mediated by absorptive capacity and moderated by organizational culture. The specific objectives include:

• Examining the role of innovation networks on knowledge diffusion;
• Analyzing the moderating role of organizational culture in the relationship between innovation networks and knowledge diffusion; and
• Evaluating the mediating role of absorptive capacity in the relationship between innovation networks and knowledge diffusion.

Through rigorous empirical analysis, this study seeks to advance academic discourse, inform management practices, and promote sustainability-oriented innovation. By uncovering the interplay among innovation networks, knowledge diffusion, organizational culture, and absorptive capacity, this research aims to help organizations unlock their full potential amid significant technological disruptions and environmental challenges. This study extends the Knowledge Spillover Theory of Entrepreneurship (KSTE) by integrating organizational culture as a key moderating factor, offering a broader theoretical framework that challenges the traditional, linear understanding of knowledge diffusion within innovation networks.

Data were collected from the Ghana Statistical Service (GSS) and the PENTSCOPE database. Moran’s I with a distance decay matrix was used to assess spatial dynamics, and all analyses were conducted using R to test the proposed theoretical model and hypotheses. This study revealed the importance of innovation networks in enhancing knowledge diffusion, supported by findings that companies leveraging innovative networks and technology experience positive impacts on knowledge diffusion. Although the specific data sources limit generalizability, they provide a basis for drawing inferences and developing new research models for the future. While this study focuses on Ghana, the findings offer valuable lessons for other emerging economies. Many face similar challenges related to innovation networks, knowledge diffusion, and absorptive capacity, making the insights broadly applicable to fostering sustainable development and entrepreneurship.

2. Literature Review and Hypotheses

2.1. Knowledge Diffusion in Innovation Networks: A KSTE Perspective

Innovation networks and knowledge diffusion are crucial for organizational performance and industry development. The importance of these networks is evident in emerging economies like Ghana, where they foster technological advancement and economic growth. Data from the Ghana Statistical Service (GSS) provide insights into local economic activities and firm-level dynamics, highlighting the boundaries of organizational economies. Such a contextual understanding sets the stage for analyzing the impacts of innovation networks within industries. Building on the insights from innovation geography and KSTE research, we propose the theoretical framework illustrated in Figure 1. We hypothesize that innovation networks are crucial for facilitating knowledge diffusion (H1). Additionally, we argue that the effectiveness of these networks in facilitating knowledge diffusion varies between organizations based on their organizational culture (H2) and absorptive capacity (H3).

2.2. Knowledge Diffusion in the KSTE

At the heart of the KSTE lies the ‘knowledge filter’ principle, which signifies obstacles and constraints that hinder the transfer from the inception of knowledge to market application. This filter is influenced by factors such as absorptive capacity. Knowledge generated in research or innovation labs does not automatically lead to commercial application; instead, it must pass through these filters. Organizational culture and absorptive capacity are critical in determining a firm’s ability to leverage external knowledge for innovation. Ref. [17] identify shared values, beliefs, and norms as critical influencers of collaboration and innovation. As outlined by [18] and further developed by [10], absorptive capacity reflects an organization’s ability to identify, assimilate, and apply external knowledge. These factors are essential for enhancing knowledge diffusion within innovation networks. Organizational culture plays a significant role in this process [17]. An obstructed filter decreases the likelihood of successful knowledge commercialization. Organizations act as intermediaries that identify opportunities, secure resources, and convert knowledge into marketable products, services, or processes [19]. However, there is still a gap in understanding how organizations can effectively overcome these knowledge filters [20].

Some perspectives suggest that larger pools of knowledge can enhance opportunities for successful spillovers [13,21]. This study extends beyond the production of knowledge to investigate the dynamics of its diffusion. We analyze how innovation networks assist organizations in navigating the knowledge filter. Leveraging upon insights from the spatial dynamics of innovation studies, our theoretical framework examines how various forms of proximity—intellectual, societal, spatial dynamics, systemic, and corporate—facilitate knowledge ties within networks [22,23]. This focus aligns with prior research indicating that knowledge networks enhance innovation performance [24,25] and that organizational clusters can accelerate knowledge spillovers [26]. Contemporary research likewise emphasizes the significance of innovation networks in driving performance [27,28] and economic development [29,30]. Collectively, these findings underscore that networks and proximity provide critical insights into knowledge diffusion among organizations and its effects on industries [31].

2.3. Innovation Networks and Knowledge Diffusion

Expanding upon previous insights, we argue that innovation networks are crucial for overcoming knowledge filters and enhancing the process of knowledge spillover, thus enabling knowledge diffusion. Robust innovation networks exhibit strong interconnectivity among members, regular cooperative efforts, and effective exchange of knowledge [32]. Such networks foster collaborative innovation and enable the swift exchange of knowledge and skills within organizations. In contrast, underdeveloped innovation networks, with limited connections and collaboration, often lead to sporadic knowledge exchange and concepts, impeding knowledge diffusion [2].

The research has shown that networks with high connectivity are crucial for effective knowledge diffusion within organizations [33]. Various factors influence knowledge diffusion, including collaborative networks, geographical proximity, and workforce characteristics. Previous studies highlight collaborative network strength [34] and geographical proximity [35] as critical determinants. This study incorporates these control variables, measuring network centrality and Euclidean distances to assess their impact. The innovative workforce’s role, as discussed by [36], further reinforces the importance of these factors in knowledge diffusion processes. A critical factor connecting innovation networks to knowledge diffusion is efficient access to advanced innovations. In organizations with robust innovation networks, the sharing and dissemination of emerging insights and innovations happen more effectively as updates on the current research and developments are circulated among network members [37,38]. This knowledge circulates within the network and reaches other organizations receptive to innovative ideas [39]. Conversely, organizations with weaker innovation networks have limited access to new ideas, making it more challenging to leverage knowledge for competitive advantage and firm growth [37].

This study adds to the existing literature by showcasing the substantial influence of innovation networks on knowledge diffusion. It contributes to the research on innovation networks [12,37,38] and knowledge diffusion [4,8,33,40,41,42] by offering empirical data from Ghanaian industries. This deepens the understanding of the role of innovation networks in enhancing knowledge diffusion within firms.

Collaborative innovation networks significantly impact new product performance by leveraging product and process innovation capabilities, supported by absorptive capacity [12]. Similarly, [38] highlight that interactions within enterprise innovation networks boost technological innovation performance but warn that overly dense networks can lead to knowledge redundancy, reducing innovation efficiency. Further elaboration on the moderating role of social capital, suggests that while connectivity within innovation networks fosters knowledge diffusion, excessive density may inhibit innovation due to saturation effects [4].

Other studies added to this discussion by examining knowledge diffusion in nascent industries [8]. They illustrate the differences between startups and established firms in driving inventions. These insights underscore that innovation networks are vital for the effective diffusion of knowledge, fostering organizational growth and competitive advantage [43].

While previous research generally supports the positive role of strong innovation networks in knowledge diffusion, caution is needed regarding the potential drawbacks of network saturation. Thus, maintaining an optimal level of connectivity is crucial to avoid diminishing returns from excessive knowledge redundancy. Contextually, we assume that solid innovation networks are essential for spreading cutting-edge innovations, providing organizations with greater opportunities to utilize new ideas and enhance their growth prospects. In contrast, weak innovation networks hinder knowledge diffusion, limiting knowledge spillovers and development opportunities. Consequently, we theorize that:

H1: 

Stronger innovation networks positively influence knowledge diffusion within industries.

2.4. The Moderating Role of Organizational Culture

Organizational culture encompasses the shared values, beliefs, assumptions, and norms that shape a firm’s behaviors, interactions, and decision-making processes. This cultural framework plays a pivotal role in influencing how firms engage in knowledge-sharing activities, both internally and externally, and it significantly impacts the effectiveness of innovation networks in fostering knowledge diffusion. In network-based innovation environments, where collaboration and external knowledge exchange are essential, organizational culture is a crucial moderator that either facilitates or constrains these processes [9].

A culture promoting openness, knowledge sharing, and adaptability enhances a firm’s ability to actively engage with external innovation networks and absorb knowledge from these partnerships [10]. Such cultures support the rapid integration of external knowledge into internal operations, enabling firms to adopt innovations more swiftly. For example, firms with continuous learning mindsets are better prepared to convert external knowledge into actionable innovation, creating a sustainable competitive advantage. In contrast, rigid or hierarchical cultures, which resist change or external input, may limit a firm’s participation in knowledge networks and restrict the diffusion of external knowledge, resulting in slower innovation cycles and underperformance. The moderating role of organizational culture is subject to debate. Strong networks may not translate into better diffusion in rigid hierarchies due to resistance to external ideas. This resistance can create knowledge silos, preventing the adequate flow and application of external knowledge, thus diminishing the potential benefits of innovation networks. Ref. [17] highlight how hierarchical cultures can obstruct knowledge-sharing processes, while [44] suggests that rigid structures limit open communication, reducing the effectiveness of innovation networks.

Furthermore, trust-based cultures emphasizing relationship-building with external partners enhance collaboration and create meaningful knowledge exchanges. These cultures enable firms to leverage innovation networks more effectively by establishing mutual trust, reducing opportunistic behavior, and improving the flow of tacit knowledge. In dynamic environments, adapting quickly to changing technologies and market conditions—fostered by a flexible culture—becomes crucial for maintaining competitiveness within innovation networks [2]. This adaptability enables firms to absorb and apply new knowledge, enhancing their innovative capacity and market positioning [7].

This discussion aligns with the network theory, which emphasizes that the success of innovation networks depends on the strength of relationships, interaction patterns, and trust among participants [2]. From this perspective, organizational culture shapes how effectively firms engage with external networks, manage knowledge flows, and foster collaboration. Cultures that encourage communication and collaboration ensure that knowledge is acquired and utilized effectively, driving competitive outcomes [17]. Therefore, organizational culture moderates the relationship between innovation networks and knowledge diffusion. When firms cultivate supportive cultures characterized by trust, openness, and adaptability, they enhance the benefits of innovation networks, facilitating greater knowledge diffusion and innovation performance. This leads to the following hypothesis:

H2: 

Organizational culture moderates the relationship between innovation networks and knowledge diffusion, enhancing this effect in culturally supportive environments.

2.5. The Mediating Role of Absorptive Capacity

Absorptive capacity refers to a firm’s ability to recognize, assimilate, transform, and apply external knowledge to create innovation and improve performance [18,45]. This capability is a vital mediator in the relationship between innovation networks and knowledge diffusion, determining how firms can convert external knowledge into actionable insights and competitive advantage. In dynamic industries, firms that cultivate absorptive capacity are better equipped to capitalize on knowledge resources from participation in innovation networks [10,44].

Innovation networks, which comprise collaborative relationships among firms, research institutions, and other stakeholders, offer significant opportunities for knowledge acquisition. However, the value derived from these networks depends on access to external knowledge and the firm’s ability to absorb and apply it effectively [46]. Absorptive capacity facilitates the acquisition and integration of tacit and explicit knowledge, ensuring that firms can adapt external knowledge to fit their unique needs [47,48]. This capacity enhances innovation outcomes by enabling firms to build on external ideas, experiment with new solutions, and create novel products or processes, thereby increasing competitiveness [49].

Moreover, firms with a higher absorptive capacity are better positioned to participate in joint R&D projects and collaborative innovation initiatives, thus further strengthening their knowledge-sharing potential [12]. In contrast, firms with a limited absorptive capacity may encounter difficulties in leveraging external partnerships, thereby missing out on critical knowledge spillovers that are essential for innovation [45]. In this context, absorptive capacity is not merely a passive outcome, but an active enabler, allowing firms to fully benefit from the knowledge available within innovation networks [50].

In the specific context of Ghana’s industries, absorptive capacity is crucial, given the rapid technological changes shaping the business landscape. Emerging economy firms must quickly adapt to evolving technologies and market demands. Those with a solid absorptive capacity can better leverage knowledge from both local and international networks, transforming it into innovative products and services that enhance competitiveness [12,49]. Firms with a robust absorptive capacity are more likely to engage in collaborative innovation, benefit from knowledge spillovers, and achieve sustainable growth in rapidly changing environments [46].

This study aligns with the Knowledge Spillover Theory of Entrepreneurship (KSTE) by demonstrating how firms leverage external knowledge to drive innovation and entrepreneurship. The KSTE posits that the diffusion of external knowledge fosters entrepreneurial activities, enabling firms to seize market opportunities. Absorptive capacity plays a vital role in this process by facilitating the transformation of external knowledge into innovations that enhance entrepreneurial performance. Therefore, developing absorptive capacity is essential for firms to benefit from knowledge diffusion through innovation networks and sustain a competitive edge.

Firms seeking to enhance their absorptive capacity should cultivate a culture of continuous learning, promote collaboration, and implement processes that facilitate the recognition, assimilation, and application of new knowledge [47]. By investing in absorptive capacity, firms can improve their ability to participate effectively in innovation networks and leverage external knowledge, thereby fostering innovation and long-term growth [7,48]. In this way, absorptive capacity serves as a mediator that strengthens the positive impact of innovation networks on knowledge diffusion, enhancing a firm’s ability to achieve sustainable innovation outcomes and gain a competitive advantage. Thus, we propose the following hypothesis:

H3: 

Absorptive capacity mediates the link between innovation networks and knowledge diffusion, with high absorptive capacity amplifying this relationship.

3. Materials and Methods

3.1. Data Collection

Given the importance of knowledge diffusion and innovation networks as central organizational factors, our empirical investigation utilized an extensive longitudinal dataset from the Ghana Statistical Service (GSS) and the PENTSCOPE database, which provides crucial economic and statistical data. The GSS data offer insights into local economic activities and firm-level dynamics, thus helping to define the boundaries of organizational economies.

Table 1. Characteristics of control variables.

Variable	Details	Mean	SD	No	Source
Knowledge diffusion	The annual number of firms registered and patented in the registrar general department (RGD) per 100 firms.	26.00	36.20	100	WIPO) Statistics Database, GIPO, ARIPO
Innovation network	The collaborative network closeness measure among patent innovators within a specific period.	0.48	2.58	100	WIPO Statistics Database, WIPO
Business turnover	A turnover rate of business entries, exits, and the number of active firms.	7.63	2.70	100	Ghana Statistical Service (GSS)
Innovative workforce	Job creation and destruction relative to the existing employment level rate.	0.24	2.23	100	Ghana Statistical Service (GSS)
Net migration ratio	The net number of migrants moving to or from a region relative to its population.	−41.22	7.12	100	Ghana Statistical Service (GSS)
R&D funds	Normalized R&D venture per corporate establishment to obtain a normalized measure.	0.88	22.73	100	GIRC Centre
Venture capital	Adjusting venture capital funds for inflation to business entities count.	0.24	3.08	100	Ghana Statistical Service (GSS)
State capacity	Relative to individual earnings and converted to a 1–5 scale, where higher scores indicate a smaller government footprint. The index is the average of these scaled components.	5.44	0.20	100	Fraser Institute’s Economic Freedom of the World Report.
Tax relief	The tax burden index is calculated using three components: (1) personal and wage tax receipts, (2) value-added tax proceeds, and (3) income from assets and various taxes. Each component is expressed as a percentage of individual earnings and converted to a 1–5 scale, where higher scores indicate lower overall taxation. The index is the average of these three scaled components.	4.80	0.64	100	Ghana Statistical Service (GSS)
Workforce flexibility	The workplace freedom metric is based on three components: (1) base salary (full-time earnings relative to per capita income); (2) public sector workforce; and (3) private sector union membership density, both (2) and (3) as percentages of total employment. Each component is scored on a 1–5 scale, where higher scores indicate fewer market distortions. The overall index is the average of these three scores.	6.20	0.10	100	Ghana Statistical Service (GSS)
Patent rate	Value patent density per 1000 inhabitants.	3.35	5.45	100	WIPO Statistics Database
Innovation grants	Adjusting for inflation and normalizing the Ghana Innovation and Research Commercialization (GIRC) Centre grants by business count.	0.22	0.74	100	GIRC Centre, Ghana Statistical Service (GSS)
Cluster strength	Concentrated or specialized employment in particular clusters is within an entity.	0.40	0.01	100	Ghana Statistical Service (GSS)
Innovation-driven workforce	The proportion of industry employment in innovation-driven workforce businesses.	1.11	0.55	100	[36]
Cluster density index	Firm concentration index, where a score of 0 represents casual location patterns and higher scores indicate more sectoral clustering.	0.04	0.03	100	[51]
Human Resources	The share of individuals aged 18+ with a bachelor’s degree or higher.	13.82	6.71	100	Ghana Statistical Service (GSS)
The firm’s patent share	The proportion of industry patents granted to organizational innovators.	73.81	2.86	100	WIPO Statistics Database
Density of large firms	The proportion of entities with 100+ employees in the firm	0.25	0.05	100	Ghana Statistical Service (GSS)
Organizational culture	Entrepreneur-to-employee ratio in non-rural industries	1.23	0.34	100	Ghana Statistical Service (GSS)
Knowledge absorptive capacity	The proportion of talent pool absorbed by industries.	1.03	0.56	100	Ghana Statistical Service (GSS)
Female board chairperson entities	Share of public companies led by female board chairs in the industry.	6.60	1.40	14	Ghana Board Diversity Index and Avance Media’s listings

provides an overview of the variables included in our analysis.

3.2. Innovation Networks

Prior research emphasizes using patent data to assess innovation networks effectively. For instance, Ref. [3] metric for internal innovator social proximity is reliable for analyzing the connections between entities within an industry. This study evaluates these networks using patent data from the World Intellectual Property Organization (WIPO) and the PENTSCOPE database. The geodesic distance between innovators helps measure network strength, providing a detailed view of industry collaboration dynamics over time.

We utilized patent data from the World Intellectual Property Organization (WIPO) to structure innovation networks on the PENTSCOPE database. These data measure the effectiveness of direct and indirect connections among innovators in an industry. More precisely, we utilized comprehensive patent application data from PENTSCOPE, which includes patents granted by the Ghana Intellectual Property Office (GIPO) and the African Regional Intellectual Property Organization (ARIPO), along with contributions from various universities and private-sector industries. PENTSCOPE provides data on granted patents and patent applications from 2000 to the present, facilitating the examination of innovation networks via a clarification algorithm applied to patent innovators, assignees, and locations. The disambiguated data were merged with additional fields to offer a comprehensive perspective on every patent file.

Over the sample period, we identified over 2500 unique firms granted patents, with approximately 500–600 patents in 2000–2009, 1200–1500 patents in 2010–2019, and 800–1000 patents in 2020-present. These patents included around 30–40% from local companies, 60–70% from foreign companies (primarily from the US, Europe, and Asia), 5–10% from research institutions and universities, and 1–5% from government agencies. Due to the complexity of patent data, potential variations in reporting, and subscription/fee access to some databases, there may be some variation in the reported figures.

To measure the strength of innovation networks, we employed [35] metric for internal innovator social proximity, as also utilized by [12]. This metric assesses the connections between entities and industries within a population. Distinct innovators within PENTSCOPE are regarded as discrete nodes, with connections established if they collaborate on a patent during the specified year. Considering all network connections, we determined the geodesical social distance (i.e., the geodesic path between innovators X and Y) for every pair of innovators within the GIPO. A combined metric of the scale and influence of the innovation network in the GIPO was derived using the following formula (Equation (1)):

$${\mathrm{I}\mathrm{N}}_{m,t}={\sum }_{j=1}^n{\sum }_{k=1}^n{\displaystyle \frac{djk}{\mathrm{n}-1}}$$

(1)

where $djk$ represents the shortest path length between innovators $j$ and $k$, and $n$ represents the total count of distinct innovators in the GIPO. By design, the innovation network score varies from 0 (indicating no collaboration) to (maximum partnership), with larger values indicating a more substantial collaboration within industries. Figure 1 depicts the variation in innovation network robustness among the GIPOs. In 2023, the median value of the GIPOs was 0.81, with a third scoring between 0.99 and 2.99. About a quarter of GIPOs attained scores exceeding 4, highlighting critical areas for innovation partnerships. The change in innovation network scores between 2000–2009 and 2020–2023 highlights temporal dynamics in network development, with increased collaboration in specific industries over time.

This measure also accounts for the evolving nature of networks, considering that innovators might engage with various partners over different periods. Consequently, we calculated innovation network robustness using two-year and three-year time frames to capture these dynamics. For example, the innovation network score for 2019 includes patents granted in 2019; the biennial score encompasses patents from 2018 to 2019, and the three-year score includes patents from 2017 to 2019.

3.3. Knowledge Diffusion

Following recent studies, we operationalized knowledge diffusion by analyzing the incidence of firms in an industry [8]. We calculated the GIPO-level ratio of firms listed on the Ghana Stock Exchange (GSE) based on the following criteria: (1) Minimum Stated Capital, (2) Profitability Record, and (3) Public Float. The firm count was assigned to the year of analysis (i.e., 2024), normalized by the total firms operating in that period. The graph in Figure 2 shows the innovation network density. We assessed industry knowledge diffusion by examining collaboration networks and entrepreneurial activity, where high proprietorship rates may suggest dynamic business environments, but do not directly equate to knowledge diffusion rates [41].

3.4. Knowledge Absorptive Capacity and Organizational Culture

• Organizational Culture: We based our scale on [17] study, which measures the influence of shared values, beliefs, and norms on behaviors, collaboration, innovation, and knowledge sharing. We considered the management approach, interaction dynamics, decision-making frameworks, workforce involvement, and openness to change. This approach has been validated by earlier studies, such as [44].
• Knowledge Absorptive Capacity: Based on [10], this scale assesses an organization’s ability to identify, integrate, transform, and apply external knowledge for innovation. Factors considered include the identification of valuable external knowledge, the ability to assimilate and transform information, and the application of this knowledge to enhance innovation and competitiveness. This aligns with the foundational work by [18].

3.5. Retrospective Analysis of Industry Policies and Patent Trends

Beyond the variables in our principal analysis, various external factors may also affect the relationship between innovation networks and knowledge diffusion. To explore this potential outcome, we conducted supplementary analyses focusing on two areas: industry policy and industry patent production. Concerning industry policy, we examined two distinct approaches: bottom-up and top-down strategies [12]. The bottom-up policy adopts a market-driven strategy to stimulate industry growth, while the top-down policy seeks to accomplish this through government initiatives and stimulus. We assessed the bottom-up policy using data from Ghana’s Local Government Performance Management System (LGPMS), which evaluates local governance performance, specifically concentrating on the administrative capacity of local government units (LGUs) in critical areas, such as service delivery, transparency, administrative capacity, and fiscal management. Our analysis uses LGPMS data to assess regional policy impacts on business activity. To evaluate industry policy from a top-down perspective, we utilized the cedi value of innovation funding awarded through the Ghana Innovation and Research Commercialization (GIRC) Centre, which supports early-stage industry ventures. Prior studies [52] show that firms receiving innovation grants are more likely to patent their innovative solutions and secure follow-on venture funding.

3.6. Control Variables

The literature on the spatial distribution of knowledge diffusion has explored a range of industry contexts, including the United States [53,54], Germany [55], China [56,57], and India [40,58]. We relied on this body of literature to inform our selection of control variables. Researchers have highlighted collaborative networks [34] and geographical proximity [35] as significant factors for knowledge diffusion within industries. We measured collaborative networks using the network centrality index from social network analysis (SNA), which captures the strength of a firm’s connections and access to knowledge flows [59,60,61]. The Euclidean distance index uses geographic coordinates to measure geographical proximity [61]. Studies also highlight the positive role of the innovative workforce [36], measured as the share of employment in innovation-driven workforce sectors. Additionally, we assessed the business ecosystem through large organization concentration and related cluster strength measures, using venture capital and research and development funding as indicators of available business investment opportunities. To capture the effects of economic activity, we incorporated the innovative workforce, business turnover, and net migration ratio. Lastly, we controlled for industry gender equality by incorporating the fraction of publicly traded firms with a female board chairperson from the ARIPO database.

3.7. Approaches

We assessed the impact of innovation networks on knowledge diffusion using the two-way fixed effects model outlined in Equation (2). In this model, KD_m,t represents our knowledge diffusion metric for GIPO $m$ at time $t$, ${IN}_{m,t}$ refers to our innovation network indicator, OC captures organizational culture, G_m,t is a matrix of GIPO-specific control variables that vary over time, while λt and μm represent the year and GIPO fixed effects, respectively, and ${\mathrm{ϵ}}_{\mathit{m}\mathit{,}\mathit{t}}$ denotes the unique error term.

$${\mathit{D}}_{\mathit{m},\mathit{t}}=\mathrm{\beta }0+\mathrm{\beta }1\mathit{I}{\mathit{N}}_{\mathit{m},\mathit{t}}+\mathrm{\beta }2\mathit{O}{\mathit{C}}_{\mathit{m},\mathit{t}}+{\mathit{v}\mathit{G}}_{\mathit{m},\mathit{t}}+\mathrm{\lambda }\mathrm{t}+\mathrm{\mu }\mathrm{m}+{\mathrm{ϵ}}_{\mathit{m},\mathit{t}}$$

(2)

The calculation begins with data collection from the PENTSCOPE database, focusing on patent records, collaboration activities, and firm-level innovation outputs. These data points provide insights into the extent of knowledge-sharing activities within each industry. Knowledge diffusion is defined by examining the incidence and impact of knowledge transfer among firms in a given industry. This involves measuring the number of collaborative patents filed between organizations within each GIPO and the ratio of firms adopting external knowledge sources for innovative outputs. The adoption is inferred from growth in patent applications, collaborative R&D projects, and citations of external patents.

To ensure comparability across different GIPOs and periods, the raw data were normalized by the total number of active firms in each industry during the specific time, $t$. This normalization process standardizes the knowledge diffusion data, creating a consistent metric across all GIPOs and timeframes. The resulting value aggregates into a composite Knowledge Diffusion Index (KDI), reflecting direct and indirect knowledge transfers, including collaborative activities and innovation outputs. Thus, $K{D}_{m,t}$ represents the overall intensity and effectiveness of knowledge diffusion within each GIPO at the time, $t$, capturing the extent of knowledge-sharing interactions and their impact on innovation within the industry. This detailed calculation thoroughly assesses the relationship between knowledge diffusion, innovation networks, and organizational factors.

The [62] test confirms that fixed effects are more appropriate than random effects. Spatial spillover effects are examined using Moran’s I with a distance-decay matrix. All analyses were conducted in R. Knowledge diffusion was assessed over three-year periods, and the base year was linked to innovation network scores.

To address potential endogeneity in the relationship between innovation networks and knowledge diffusion, we employed carefully selected instrumental variables (IVs) that met the relevance and homogeneity criteria. Patent data availability reflects long-term innovation potential rather than short-term economic variations, and government innovation funding captures external policy shocks independent of firm-specific behavior. Thus, patent data availability is an instrument for the strength of innovation networks.

The validity of the instrumental variables (IVs) was rigorously tested to ensure their reliability in mitigating potential biases. The Sargan test results confirm no over-identification issues across most models, with p-values ranging from 0.047 to 0.850, indicating that the instruments are valid. For example, the Sargan test p-values for Model 1 and Model 3 were 0.710 and 0.750, respectively, supporting instrument validity. Additionally, the first-stage Cragg–Donald F-statistics ranged from 5.08 to 78.47, with values such as 24.38 (Model 1) and 66.78 (Model 2) demonstrating strong instrument relevance. These robust tests confirm that our IV approach effectively isolates the causal relationship between innovation networks and knowledge diffusion, minimizing endogeneity concerns.

4. Results

4.1. Key Findings

Table 2. Fixed effects parameter results.

Variable	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Innovation Network (IN)	0.200 ** (0.043)	−2.065 (1.036)
Innovation Network (IN), 2-year			0.032 *** (0.021)	−0.707 (0.114)
Innovation Network (IN), 3-year					0.074 *** (0.011)	−0.138 (0.788)
Organizational culture	2.143 *** (0.763)	1.801 *** (0.814)	2.157 *** (0.760)	1.851 *** (0.800)	2.168 *** (0.760)	2.008 *** (0.803)
Organizational culture * IN		0.064 (0.002)
Organizational culture * IN, 2-year				0.012 (0.053)
Organizational culture * IN 3-year						0.060 (0.037)
Knowledge absorptive capacity	2.140 *** (0.813)	1.811 *** (0.754)	2.050 *** (0.734)	1.661 *** (0.768)	2.054 *** (0.751)	2.034 *** (0.801)
Knowledge absorptive capacity * IN		0.153 (0.012)
Knowledge absorptive capacity * IN, 2-year				0.011 (0.142)
Knowledge absorptive capacity * IN, 3-year						0.151 (0.126)
No	100	100	100	100	100	100
Venture capital	0.214 (0.205)	0.208 (0.204)	0.215 (0.205)	0.211 (0.204)	0.213 (0.205)	0.210 (0.203)
Business turnover	−0.705 (0.520)	−0.715 (0.517)	−0.706 (0.520)	−0.726 (0.518)	−0.708 (0.520)	−0.731 (0.518)
Innovative workforce	0.214 * (0.052)	0.202 * (0.052)	0.215 * (0.052)	0.202 * (0.052)	0.216 * (0.052)	0.201 * (0.052)
R&D funding	−0.051 (0.035)	−0.055 (0.035)	−0.051 (0.035)	−0.055 (0.035)	−0.051 (0.035)	−0.054 (0.035)
Net migration ratio	−0.027 *** (0.003)	−0.027 *** (0.003)	−0.027 *** (0.003)	−0.026 *** (0.003)	−0.027 *** (0.003)	−0.026 *** (0.003)
State capacity	−0.304 (1.560)	−0.323 (1.538)	−0.306 (1.565)	−0.357 (1.548)	−0.278 (1.568)	−0.386 (1.561)
Tax relief	7.015 * (3.300)	7.300 * (3.332)	6.884 * (3.304)	7.353 * (3.323)	6.880 * (3.307)	7.352 * (3.310)
Workforce flexibility	−0.385 (1.815)	−0.560 (1.856)	−0.438 (1.817)	−0.648 (1.865)	−0.441 (1.815)	−0.670 (1.866)
Patent rate	0.483 (0.261)	0.423 (0.260)	0.484 (0.261)	0.417 (0.260)	0.507 (0.261)	0.430 (0.261)
Innovation grants	0.830 (0.808)	0.826 (0.810)	0.808 (0.813)	0.827 (0.814)	0.821 (0.810)	0.846 (0.810)
Cluster strength	−11.603 **	−11.507 **	−11.700 **	−11.688 **	−11.817 **	−11.830 **
Innovation-driven workforce	0.201 (1.845)	0.187 (1.856)	0.138 (1.840)	0.114 (1.851)	0.117 (1.841)	0.068 (1.848)
Cluster density index	−1.534 * (0.270)	−1.570 * (0.272)	−1.520 * (0.270)	−1.546 * (0.276)	−1.520 * (0.270)	−1.537 * (0.278)
Human resources	1.043 * (0.003)	1.065 * (0.017)	1.067 * (0.002)	1.087 * (0.013)	1.072 * (0.004)	1.007 * (0.014)
The firm’s patent share	0.001 (0.028)	0.003 (0.028)	0.020 (0.028)	0.004 (0.028)	0.001 (0.028)	0.004 (0.028)
Density of large firms	−4.006 (11.635)	−4.081 (11.488)	−4.048 (11.647)	−4.226 (11.548)	−4.030 (11.658)	−4.280 (11.584)
Female board chairperson entities	0.870 ** (0.688)	0.806 ** (0.670)	0.877 ** (0.688)	0.784 ** (0.666)	0.877 ** (0.688)	0.775 ** (0.666)
Adjusted R-squared	0.636	0.636	0.636	0.636	0.636	0.636
Hausman test	66.67 ***	41.61 ***	68.36 ***	37.6 ***	70.08 ***	37.25 ***
Moran’s I p-value	0.103	0.100	0.104	0.115	0.088	0.113
GIPO constant effects	Yes	Yes	Yes	Yes	Yes	Yes
Annual constant effects	Yes	Yes	Yes	Yes	Yes	Yes
p(IN = IN × organizational culture = 0)	0.012		0.011		0.021
p(IN = IN × absorptive capacity = 0)		0.028		0.003		0.001

Note: Innovation network is the independent variable and knowledge diffusion is the dependent variable. All models are computed using ordinary least squares (OLS) regression, including fixed effects for the GIPO and years and an intercept. Errors are grouped at the industry level. The p-values shown are from combined statistical tests of the innovation network’s dominant effect and the specified synergistic effect. A Hausman test compared fixed effects (FEs) and casual effects (CEs) stipulations. The p-value from Moran’s I statistic assesses the remaining spatial correlation using a distance-decay weighting matrix. For detailed characteristics of the variables, refer to Table 1. Significance levels are indicated as follows: *** p < 0.01, ** p < 0.05, and * p < 0.10.

presents our key findings. Model 1 serves as the foundational estimate of the impact of innovation networks on knowledge diffusion, controlling for several variables. As hypothesized (H1), innovation networks are strongly positively correlated with knowledge diffusion; for each unit increase in the innovation network measures, there is a corresponding increase of 0.200 additional knowledge diffusion firms per entity (β = 0.200; p = 0.038; 95% CI = [0.000, 0.508]). Furthermore, organizational culture has a significant direct positive effect on knowledge diffusion, with a one-percentage point increase in organizational culture leading to 2.14 additional knowledge diffusion businesses per location (β = 2.143; p < 0.001; 95% CI = [0.386, 4.000]). The absorptive capacity measure is also significant (β = 1.032; p < 0.001; 95% CI = [0.274, 1.790]).

Control variables show that industries with higher innovative workforce rates, business turnover rates, a higher percentage of female board chairs, more collaborative networks, and higher patenting rates are associated with greater knowledge diffusion, while industries with a higher net migration ratio, greater cluster strength, and more geographical proximity exhibit less diffusion. Other controls are not significant at the 10% level.

Model 2 adds the interaction between innovation networks and organizational culture to test H2.

The combined assessment of the total marginal impact of innovation networks (labeled p(IN + IN × organizational culture = 0)) is 0.028, indicating that the combined effects are statistically significant.

The main impact of innovation networks is negative (β = −2.056; p = 0.047; 95% CI = [−6.281, 0.126]), but their interaction with organizational culture is positive (β = 0.064; p = 0.017; 95% CI = [−0.041, 0.302]).

Absorptive capacity continues to mediate positively between innovation networks and knowledge diffusion.

Models 3 to 6 examine different time windows for the innovation network measures. In Model 3, leveraging a two-year patent period, a one-unit surge in the innovation network measure correlates with an additional 0.032 firms with enhanced absorptive capacity per location (β = 0.032; p < 0.001; 95% CI = [0.016, 0.048]), while in Model 5, using a three-year window, a unit increase results in an additional 0.063 firms with a solid organizational culture (β = 0.063; p < 0.001; 95% CI = [0.018, 0.108]). The effects weaken over time, suggesting that local knowledge diffusion plays a more prominent role in shaping knowledge networks.

Finally, interaction regressions across Models 4 and 6 are consistent with those in Model 2. While absorptive capacity is associated with higher knowledge diffusion, the effect is imprecisely measured, and no residual spatial correlation was detected through Moran’s I test. Figure 3 illustrates that the marginal impact of innovation networks on absorptive capacity is more robust in industries with a higher organizational culture, confirming H2 and H3. The results suggest that innovation networks significantly benefit industries with an above-average organizational culture.

4.2. Validation Checks Employing Instrumental Variables

A key issue with our key findings is the possible endogeneity of the innovation network measures. In this segment, we evaluate the reliability of the findings shown in Table 2 by applying two alternative instrumental variable (IV) methods. Each method relies on distinct assumptions, providing a valuable sensitivity analysis for the OLS estimates, which presume that the innovation network scores are exogenous.

The initial IV method employs external instruments to capture exogenous variations in the innovation network score, facilitating causal inference. An effective external instrument should exhibit a strong correlation with the innovation network score and should only influence knowledge diffusion indirectly through its impact on innovation networks. The second characteristic, the exclusion restriction, cannot be empirically tested. We utilize innovation network data from 10 and 25 years earlier to instrument the innovation network scores. For example, the external instrument for the 2000 innovation network score is based on patents in 2010–2023 (14-year lag) and 2000–2009 (10-year lag). The instruments for later years adjust accordingly. The persistence of innovation culture within industries [17] suggests past network strength correlates with current innovation networks. However, for the exclusion to be valid, the past scores must not affect current knowledge diffusion.

Given that patents peak in citations soon after being awarded [63] and most patent holders still require long patenting careers [64], it is plausible that current knowledge diffusion is unrelated to past innovation networks. Nonetheless, the exclusion restriction is violated if past networks influence current knowledge diffusion. To address this, we employ an alternative instrumental variable method utilizing instruments derived from the dataset through higher-order moment constraints [65]. These instruments are drawn from collaborative networks, the innovative workforce, and net migration ratios.

Table 3. Instrumental variable approximations.

Variable	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Panel X: outcomes via external instruments
Innovation Network (IN)	0.217 ** (0.048)	−1.705 (2.002)
Innovation Network (IN), 2-year			0.082 * (0.035)	−0.402 (0.357)
Innovation Network (IN), 3-year					0.038 * (0.014)	−0.044 (0.875)
Organizational culture	2.253 *** (0.757)	1.806 *** (0.868)	2.162 *** (0.760)	2.006 *** (0.822)	2.170 *** (0.760)	2.043 *** (0.808)
Organizational culture * IN		0.056 (0.055)
Organizational culture * IN, 2-year				0.074 (0.068)
Organizational culture * IN 3-year						0.048 (0.042)
Knowledge absorptive capacity	2.243 *** (0.630)	1.810 *** (0.738)	2.252 *** (0.571)	2.103 *** (0.832)	2.271 *** (0.760)	2.061 *** (0.7010)
Knowledge absorptive capacity * IN	0.155 (0.144)
Knowledge absorptive capacity * IN, 2-year			0.165 (0.157)
Knowledge absorptive capacity * IN, 3-year						0.139 (0.143)
Strong Cragg–Donald	24.38	61.76	78.47	19.15	14.83	21.65
Sargan p-value	0.710	0.52	0.750	0.526	0.77	0.503
Panel Y: outcomes via Quiroga, (2021) generated instruments
Innovation Network (IN)	0.270 ** (0.048)	5.207 (5.524)
Innovation Network (IN), 2-year			0.038 *** (0.031)	2.370 (1.807)
Innovation Network (IN), 3-year					0.071 *** (0.014)	1.632 (1.324)
Organizational culture	2.137 *** (0.760)	2.742 *** (0.058)	2.156 *** (0.758)	2.680 *** (0.018)	2.168 *** (0.760)	2.710 *** (0.025)
Organizational culture * IN		−0.207 (0.242)
Organizational culture * IN, 2-year				−0.064 (0.042)
Organizational culture * IN, 3-year						−0.030 (0.018)
Knowledge absorptive capacity	2.134 *** (0.760)	2.742 *** (0.155)	2.166 *** (0.750)	2.449 *** (0.116)	2.046 *** (0.761)	2.711 *** (0.123)
Knowledge absorptive capacity * IN	−0.407 (0.412)
Knowledge absorptive capacity * IN, 2-year			−0.024 (0.143)
Knowledge absorptive capacity * IN, 3-year					−0.022 (0.105)
Strong Cragg–Donald	22.56	66.78	70.22	5.08	68.86	17.35
Sargan p-value	0.466	0.047	0.850	0.850	0.162	0.440

Note: The independent variable in the model is the innovation network, while the dependent variable is knowledge diffusion. All models are computed using ordinary least squares (OLS) regression, with fixed effects for the GIPO and years and an intercept term. Errors are grouped at the industry level. The strength of the instruments used in the model is evaluated by the Cragg–Donald statistic. The Sargan p-values derive from tests of instrument validity, ensuring the instruments are appropriately exogenous. For detailed descriptions of the variables, refer to Table 1. Control variables are excluded for conciseness. Statistical levels are denoted as *** p < 0.01, ** p < 0.05, and * p < 0.10.

summarizes the IV results. Panel X shows estimates using external instruments, while Panel Y uses Quiroga’s instruments. The Sargan overidentification test does not reject the null hypothesis in all models, suggesting that the instruments used are appropriate and valid. The Cragg–Donald statistics show robust instruments, especially in Panel X, with minimum F-statistics well above the threshold of 104.7 [66]. Panel Y, however, shows weaker instruments, particularly in interaction models.

Despite these differences, the instrumental variable estimates presented in Panel X align closely with the OLS results. For instance, an increment of one unit in the innovation network score results in 0.200 (p = 0.038) additional companies in Table 2, compared to 0.217 (p = 0.033) when employing external instruments in Panel X and 0.270 (p = 0.010) when utilizing data-based instruments in Panel X. While the OLS and IV results support H2 and H3, the IV results using data-generated instruments are less precise, especially for engagement frameworks. Consequently, the IV approximations from Panel X, utilizing external instruments, are more credible for identifying causal effects. The marginal effects of innovation networks on firms in industries with higher knowledge absorptive capacity remain significant in Panel X but become less statistically supported in Panel Y.

4.3. Preliminary Investigations

As mentioned above, we perform a retrospective analysis to investigate how bottom-up and top-down organizational policies moderate industry knowledge synthesis. Bottom-up strategies encompass pro-market mechanisms (e.g., state capacity, tax relief, and workforce flexibility), while top-down strategies are assessed via innovation grants. Patent rates represent industry knowledge synthesis. To simplify this, we display the distinct marginal effects of individual moderators across distinct panels in Figure 4. We estimate our exploratory models using OLS because of its alignment with IV approaches, applying current innovation network scores. In general, insufficient evidence indicates that either type of strategy substantially moderates the link between innovation networks and knowledge diffusion. Innovation grants, in particular, show little effect. However, more robust innovation networks show a modest advantage in industries with increased patent filings, although the economic impact remains trivial. Interestingly, industries with above-average tax relief (0.5 SD above the mean) exhibit a significantly more robust relationship between innovation networks and knowledge diffusion than those with moderate or lower incentives.

5. Discussion

This study provides strong empirical evidence on how innovation networks foster knowledge diffusion, emphasizing absorptive capacity’s mediating role and organizational culture’s moderating role. Grounded in the Knowledge Spillover Theory of Entrepreneurship (KSTE), we find that firms can convert external knowledge into sustainable innovation through active participation in collaborative networks.

The results align with prior studies showing that innovation networks enable the flow of ideas and resources between firms, allowing them to access critical external knowledge. Specifically, firms embedded in robust networks are more effective at integrating and applying acquired knowledge, thus reinforcing their innovation capacity [2].

This study also confirms that absorptive capacity is vital for firms to recognize, assimilate, and exploit new information [18,45]. Organizations with more robust absorptive capacities are better equipped to turn external knowledge into practical innovations, enhancing competitiveness [47,48].

Moreover, organizational culture plays a moderating role in knowledge diffusion. Firms with collaborative and adaptive cultures are more likely to integrate external knowledge effectively [17]. In contrast, rigid or hierarchical cultures hinder knowledge sharing and innovation. Cultures that promote trust also strengthen collaboration within networks, improving firms’ ability to absorb and diffuse knowledge.

Additionally, this study highlights the dynamic nature of knowledge diffusion within networks. The impact of innovation networks is most substantial in the short term, but it typically weakens over time. This temporal complexity calls for future studies to examine how firms can sustain the effectiveness of these networks over longer periods. The challenges and opportunities identified in this study are expected to benefit Ghana. Many emerging economies share structural similarities, such as developing innovation infrastructures and varying levels of absorptive capacity. Therefore, the observed mechanisms—particularly the roles of innovation networks and organizational culture—can inform strategies in other regions aiming to enhance knowledge diffusion and technological advancement.

Our findings demonstrate that strong innovation networks significantly enhance knowledge diffusion, a phenomenon not unique to Ghana but applicable to emerging economies worldwide. These results align with global innovation frameworks and suggest that fostering robust networks could be a strategic priority for policymakers aiming to accelerate technological advancement and economic growth. For instance, similar policies could be applied in other developing regions, such as Southeast Asia or Latin America, to stimulate innovation through enhanced knowledge-sharing platforms. Furthermore, multinational firms operating in diverse cultural environments could leverage these insights to tailor their innovation strategies, emphasizing the role of organizational culture and absorptive capacity in facilitating effective knowledge diffusion.

5.1. Theoretical Contributions

This research significantly advances continuity in the academic discourse by integrating the Knowledge Spillover Theory of Entrepreneurship (KSTE) with network theory. It elucidates how firms exploit external knowledge via innovation networks to sustain knowledge diffusion across industries [45]. Our findings indicate that absorptive capacity is a crucial mediator of a sustainable relationship between innovation networks and knowledge diffusion, enriching the current theoretical frameworks [50].

Additionally, we emphasize the pivotal moderating role of organizational culture, which influences how effectively firms can sustain harnessing external knowledge. Specifically, organizations characterized by adaptive and collaborative cultures are better positioned to integrate new information, whereas rigid cultures may obstruct and challenge the continuity of knowledge sharing [2]. This research sheds light on the interplay between internal organizational practices and external knowledge flows, offering new insights into optimizing innovation processes and sustaining competitive advantages through strategic knowledge management.

Leveraging insights from spatial dynamics and the KSTE, we develop a theoretical framework to analyze the influence of innovation networks on knowledge diffusion. This framework underscores the contingent effects of organizational culture and absorptive capacity. Furthermore, we investigate the roles of public policy and patent production in shaping this relationship. Employing robust empirical techniques to address potential biases, our results validate the nuanced role of innovation networks in facilitating long-lasting knowledge diffusion.

We contribute to the KSTE by extending its applicability to sustainable diffusion mechanisms and demonstrating its capacity to overcome the ‘knowledge filter,’ mainly through the lens of spatial factors derived from innovation geography literature. Our analysis reveals that knowledge diffusion and production are critical for creating growth-enhancing spillovers while also addressing the temporal complexity of these interactions. Notably, we find that the impact of innovation networks is most pronounced in contemporary settings, suggesting a decline in effectiveness over time. This dynamic nature of knowledge diffusion highlights the necessity for future research to explore time-dependent effects within innovation networks.

Our findings demonstrate that the mechanisms driving innovation networks and knowledge diffusion are not limited to the specific context of Ghana, but can be applied broadly across various emerging and developed economies for sustainable knowledge diffusion.

In particular, policymakers worldwide can leverage these insights to design strategies that accelerate technological advancement for sustainable economic growth through enhanced knowledge-sharing platforms.

This aligns with global innovation frameworks that emphasize the importance of networked collaboration in driving sustainable innovation.

The implications of our research extend beyond national borders, with particular relevance for regions like Southeast Asia and Latin America, where innovation ecosystems are developing rapidly but may face sustainability challenges relating to absorptive capacity, organizational culture, and network strength. For multinational firms operating in diverse cultural environments, our findings suggest that tailoring innovation strategies to local contexts—by focusing on organizational culture and building robust innovation networks—can facilitate more effective knowledge diffusion.

Moreover, this research contributes to the global discourse on public policy by illustrating how government strategies—such as tax incentives, innovation grants, and institutional frameworks—can be designed to foster and sustain knowledge diffusion and innovation on a broader scale. As global innovation networks become increasingly interconnected, our results provide a timely contribution to understanding how cross-border collaboration, organizational practices, and national policies shape the future of innovation and entrepreneurship worldwide.

5.2. Management Implications

This study offers essential insights for entrepreneurs, managers, and policymakers. It provides actionable recommendations that can enhance organizational effectiveness and foster innovation.

Implications for Entrepreneurs and Managers: We highlight two critical strategic considerations for entrepreneurs. First, those aiming for ambitious growth should prioritize their business location and focus on the strength of local innovation networks and their firm’s absorptive capabilities. This principle is not limited to emerging economies; it applies globally, especially in rapidly evolving sectors, such as technology, healthcare, and green energy. Engaging in collaborative networks is vital for accessing external knowledge and fostering continued innovation [12].

Second, expanding networks beyond immediate peers to include international partners and diverse sectors can significantly enhance exposure to emerging trends. This is particularly relevant in today’s globalized economy, where cross-border knowledge exchange can offer competitive advantages. To capitalize on these opportunities, small startups or large multinationals must invest in processes that enhance knowledge assimilation, such as robust R&D systems, comprehensive staff training, and internal learning mechanisms.

Cultural and Structural Adaptability: Our findings suggest that adaptive and collaborative organizational cultures are universally beneficial. These insights are relevant to emerging economies like Ghana, organizations operating in culturally diverse environments, and multinational corporations. Managers in global firms should foster open cultures that encourage experimentation and knowledge sharing across borders and functions [2,17].

Global Policy Implications: Our research underscores the importance of nuanced, context-specific approaches to fostering innovation. While top-down initiatives, such as innovation grants, are shared globally, their effectiveness varies significantly based on the strength of local innovation networks. Our findings indicate that more than simply increasing financial incentives may be necessary, especially in developed economies where innovation ecosystems are mature but may face diminishing returns.

Conversely, bottom-up approaches, such as tax incentives that alleviate financial burdens, show promise across different economic contexts. Countries with substantial innovation ecosystems, like Germany and Japan, could benefit from these findings by tailoring tax policies to support industries with collaboratively solid networks. Similarly, developing nations can leverage these insights to design policies encouraging sustainable, network-driven entrepreneurship rather than subsidy-dependent ventures.

Implications for International Collaboration: Our findings also highlight the importance of international knowledge networks in a globalized economy. Policymakers and international organizations (e.g., the OECD and UN) should consider fostering cross-border innovation networks to enhance global knowledge diffusion. For instance, public policies promoting international R&D collaborations or digital knowledge-sharing platforms could amplify the benefits identified in this study.

We aim to provide actionable insights to inform strategic decisions in diverse organizational and policy environments by framing our findings in a broader global and cross-industry context. Future research should explore how these principles apply to emerging trends such as AI-driven innovation, remote collaboration, and sustainable development initiatives.

5.3. Conclusions

This study provides compelling evidence of how innovation networks significantly foster knowledge diffusion within firms, highlighting the pivotal roles of absorptive capacity and organizational culture. Grounded in the Knowledge Spillover Theory of Entrepreneurship (KSTE), our findings demonstrate that firms embedded in robust innovation networks can effectively convert external knowledge into sustainable innovations, thereby gaining a competitive edge. The mediating role of absorptive capacity underscores the importance of firms’ ability to recognize, assimilate, and apply new information. At the same time, an adaptive organizational culture enhances this process by promoting collaboration and knowledge sharing.

Our research offers significant theoretical contributions by extending the KSTE framework and integrating insights from network theory. This integration sheds light on how firms exploit external knowledge and the moderating role of organizational culture. By addressing the dynamic nature of knowledge diffusion and the contingent effects of network structure and absorptive capacity, we provide a nuanced understanding of the complexities involved in innovation processes. The insights from this research extend beyond the Ghanaian context, offering a framework for other emerging economies to leverage innovation networks for knowledge diffusion and sustainable innovation. Policymakers and firms in similar markets can adopt these strategies to overcome structural challenges and drive long-term growth.

From a practical standpoint, the study offers valuable implications for entrepreneurs, managers, and policymakers. Firms should strategically engage in collaborative networks and foster organizational cultures that support knowledge diffusion. Policymakers must recognize the importance of creating conducive environments that promote network formation and sustain innovation ecosystems. Future research should build on these insights by exploring the evolving dynamics of innovation networks and examining the impact of inter-innovation and intra-innovation networks over time. In an increasingly interconnected world, harnessing the potential of innovation networks remains essential for driving sustainable growth and long-term success.

5.4. Limitations and Future Research

While this study provides valuable insights into the role of innovation networks in knowledge diffusion, it has certain limitations that warrant consideration for future research. First, our research primarily relies on data from the PENTSCOPE database, which focuses exclusively on patents granted in Ghana and nearby regions. This localized perspective may limit the generalizability of our findings. Future studies could enhance understanding of knowledge diffusion mechanisms by incorporating qualitative methods across diverse contexts and industries. For instance, investigating the dynamics of knowledge diffusion in different geographical settings could yield insights into how varying cultural and economic environments influence innovation networks. Furthermore, exploring how innovation networks evolve and considering additional moderating factors, such as policy environments and technological infrastructure, could deepen our understanding of their impact on knowledge diffusion.

Preliminary analyses indicated that these factors did not significantly affect our main results. At the same time, we examined inter-innovation networks—specifically, the external social proximity of connections between industry innovators and those in different sectors within Ghana. This suggests that geographic and interpersonal connections remain crucial for facilitating knowledge spillovers [7]. Future research should investigate whether inter-innovation networks complement or substitute intra-innovation networks in promoting knowledge spillover and entrepreneurial activities. Additionally, exploring the dynamics of specific types of knowledge generation regarding cross-geographic collaborations could provide further clarity on how knowledge flows within innovative environments [23,67].

Lastly, our evaluation of innovation network strength through the shortest path length between innovators primarily addressed direct and indirect connections relevant to knowledge diffusion [61]. However, we should have accounted for varying network connectivity conditions, such as the influence of tightly knit communities or the roles of bridging connections [68]. Moreover, the need for more significant differentiation between internal and external organizational networks or joint ventures may have limited our analysis [69,70]. Future research could extend this framework by incorporating these dimensions, providing a more nuanced understanding of how diverse network structures affect knowledge diffusion.