Exploring a Sustainable Pathway Towards Enhancing National Innovation Capacity from an Empirical Analysis

Abstract

Innovation is a strategic driver of sustainable competitive advantage and long-term economic growth. This study proposes an empirical framework to support the sustained development of national innovation capacity by examining key enabling factors. Drawing on an extensive review of the literature, the research investigates the interrelationships among governmental support (GS), innovation agents (IA), university–industry R&D collaborations (UIRD), and innovation cluster development (ICD), and their influence on two critical innovation outcomes, knowledge creation (KC) and knowledge diffusion (KD). Using panel data from G7 countries spanning 2008 to 2018, sourced from international organizations such as the World Bank, the World Intellectual Property Organization, and the World Economic Forum, the study applies regression analysis to test the proposed conceptual model. Results highlight the foundational role of GS in providing a balanced framework to foster collaborative networks among IA and enhancing the effectiveness of UIRD. Furthermore, IA emerges as a pivotal actor in advancing innovation efforts, while the development of innovation clusters is shown to selectively enhance specific innovation outcomes. These findings offer theoretical and practical contributions for policymakers, researchers, and stakeholders aiming to design supportive ecosystems that strengthen sustainable national innovation capacity.

1. Introduction

The capacity of a country to innovate serves as a scale to assess its intellectual, economic, and social well-being and success. From a dynamic capacity approach, innovation can be defined as a continuous ability to transform knowledge and ideas into new processes, systems, and products to benefit an organization and its stakeholders [1]. Furthermore, innovation capacity is among the most important knowledge-based intangible assets for competing, surviving, and sustaining long-term undertakings [2,3]. Thus, innovation plays a significant role in the competitive capacity of a nation [4].

Innovation encompasses techniques and approaches for developing new products and services, markets, and redesigning production and processes [5]. By adopting a proactive approach towards innovation, developed countries have enhanced their value chain, stimulated their economies, and strengthened their participation in global markets, positioning them as leaders in the competitive landscape [6,7,8]. For developing countries, it might improve their competitiveness and growth by implementing innovative research and development (R&D) policies and practices [5,9,10]. As a dynamic capability, innovation enhances a competitive and sustainable advantage by providing organizations with creative responses to customers’ demand and the competition’s pressure [11,12]. This capacity enables SMEs to create value and knowledge and implement disruptive technology as a means to reach sustainability [2,13,14]. In light of this, the United Nations Development Program includes innovation in the ninth sustainable goal as a critical driver of sustainability and economic growth [15]. It promotes sustainable industries, scientific research, and technological progress by boosting the search for sustainable solutions to economic, social, and environmental challenges while sustaining development [16,17,18]. Additionally, innovation capability stimulates the design of sustainable products and services and the implementation of eco-efficient and cleaner production practices to create sustainable value [19,20,21,22,23].

Successful innovation emerges from a range of potential capabilities and a combination of forward-thinking ideas [24]. Moreover, this capacity involves several approaches such as a national perspective [25,26], regional perspective [27,28], organizational perspective [29,30], among others. Thus, several determinant, multi-faceted constructs and mechanisms have been identified as related to the development of this capability [26,31,32]. Moreover, various mechanisms have been developed and proposed to examine the impact, role, and performance of several determinants encompassed by this capacity across many countries from different approaches [26,33,34,35].

Given the importance and competitive potential of innovation capacity, policymakers and stakeholders in both the private and public sectors must develop a comprehensive policy framework. This framework should effectively manage innovation determinants, enabling organizations and the national innovation system to achieve a higher level of innovation, and as a result, develop a sustainable competitive advantage for themselves and the overall national innovation system. Despite the recognized key role played by innovation capacity to enhance national sustainability, competitiveness, and economic growth, this field of study remains fragmented due to the complexity and heterogeneity of its determinants. Furthermore, prior studies frequently focus on isolated indicators or constructs, without capturing the interdependent nature of this capacity across time and national contexts.

This relatively new field of study has attracted the attention of scholars and practitioners who have conducted various studies to deepen the understanding of the performance, impact, and role of several determinants of the innovation capacity [28,36,37,38,39,40,41,42]. Nevertheless, the research in this field is difficult to trace and track as several determinants have been adjusted, redefined, excluded, or incorporated over the years, making it hard to identify their behavior and real impact on innovation capacity [4,9,43,44,45]. Hence, few studies examine a broad set of determinants to analyze their relationships and effects on innovation capacity due to the complexity of the multi-faceted nature of this capacity [25,27,46,47,48,49]. Also, including every determinant proposed by various entities and literature could be overwhelming and ineffective when designing a comprehensive policy framework to promote innovation capacity and R&D.

This context leads to the following research question: What is an effective approach for a country to establish an effective framework that develops and sustains its innovation capacity to achieve innovative outcomes? To address this question, this paper presents an exploratory analysis for a pathway to enhance innovation capacity by analyzing empirical evidence from seven countries. From an extensive literature review, six key constructs were identified to set the pathway to be empirically assessed. This research offers three novel contributions. First, this study identifies and integrates six theoretically grounded constructs reflecting the multifaceted nature of innovation capacity. These constructs function as spanning inputs, enabling conditions, or innovation outputs in a cohesive framework. By assessing the proposed pathway across G7 countries, this study fills a significant gap in comparative innovation policy research. Second, this study introduced a novel harmonization technique using k-means clustering to standardize heterogeneous innovation determinants on a common ordinal scale. This provides balanced empirical rigor with interpretability, allowing effective cross-construct comparison. Finally, this research adopted a regression-based exploratory pathway to evaluate how the identified constructs impact innovation outputs, thus offering a practical framework, grounded in empirical evidence, to develop policies and strategies that enhance innovation capacity, thereby sustaining the competitiveness of a country.

The rest of this work is structured as follows. Section 2 presents a theoretical background and hypothesis development. Section 3 displays the empirical design. Section 4 reports empirical findings. Section 5 provides a discussion. Section 6 presents the conclusions and limitations of this work.

2. Theoretical Background and Hypothesis Development

Innovation is a key factor in economic development [50,51,52], particularly in recent decades, due to technological innovations leading to rapid changes in the socioeconomic global landscape [53,54,55,56]. The innovation capacity of a country, as an economic and political entity, refers to its long-term ability to generate and profit from novel, unprecedented technologies [26]. Furthermore, every nation displays distinct innovation grades as outcomes from “cross-country innovation policy” and “economic geography” distinctions [9,57,58,59]. Industry decision-makers, universities, government agencies, policymakers, investors, and researchers, among others, interact with each other to create innovative solutions, adding to the complexity of the national innovation system [60,61,62]. As a system, every action and event taking place between its agents and variables will impact the innovation determinants affecting the performance of the innovation capacity of such a system [63,64,65]. Assessing the performance of a complex system is a challenging and intricate endeavor [18,26,66,67,68]. Additionally, “[assessment] is not contradictory within the process of innovation, but can rather be used as a tool for developing innovation capability” [69] (p. 163). Hence, it is necessary to identify and monitor key indices and indicators for assessing the performance of this capacity, in the context of regional economic development and transitions, as a source of national competitiveness [44,70].

Several scholars and organizations have proposed different mechanisms to assess the innovation capacity of a nation [71,72,73]. For example, the European Commission conducts the European Innovation Scoreboard (EIS) [74], the World Intellectual Property Organization developed the Global Innovation Index (GII) published yearly [44], Furman et al. [26] proposed a framework to explore the determinants of national innovation capacity (NIC), and the World Economic Forum presented the Global Competitive Index (GCI 4.0) [4]. Each of these proposes a particular set of methods, determinants, and interpretations to evaluate innovation at a national level. However, there are similarities among them. Thus, we framed this research on the GII, NIC, and GCI 4.0.

2.1. Governmental Support

Innovation is a risky and proactive undertaking that demands skilled human capital, financial resources, and technology investments [75,76,77]. Its development largely depends on the extent and strength of a nation’s common innovation infrastructure [4,26,44]. This infrastructure corresponds to the most relevant policy and investment decisions supporting innovative efforts that have a broad effect throughout an economy [26,58]. Thus, the policy framework of a country’s government to promote and protect intellectual property (IP) rights and innovation (GS_IP_PROTC) will have an impact on investing in innovation [78,79]. The governmental approach to IP protection will further configure the innovation national system. Indeed, previous research found a positive impact of public policies protecting IP on investing in R&D and innovation endeavors [52,80,81,82,83]. Holmes et al. [82] found a strong and positive relationship between the IP protection framework and foreign R&D investment decisions, i.e., strong levels of protection impacted positively foreign R&D investment and vice versa. Furthermore, Nhemachena et al. [78] identified a strong linkage between releasing new varieties and wheat productivity in plant breeding and strengthening IP protection systems. Nevertheless, the authors also found that invigorating IP protection systems is not enough incentive to promote investments in innovation for all scenarios. Analogous findings point to the relevance of protecting and promoting IP and R&D and their relationship to improve innovation capacity [83,84,85].

The openness of a country to international trade and foreign direct investment (FDI) constitutes a critical governmental policy approach that will limit or stimulate its innovation capacity (GS_OPEN_INV). It encompasses the development of international policies and programs to reach regional, national, and local advantages from integration, spillovers, and trade [9,40,86,87,88,89,90,91,92,93]. For example, Wu et al. [92] examined 80 countries in the years 1981–2010, and found that both FDI and high-tech related international export contribute to the ability of emerging countries to create “cutting-edge technologies”, although the same effect was not found for leading innovator countries. Regarding the latter countries, the authors also identify a high correlation between IP rights protection and international patenting activities and a positive impact on national innovative capacity. Allard and Williams [94] observed two roads to innovation: one domestic orientation (i.e., control of corruption, financial inclusion, and education) adopted by higher-income countries, and an international orientation (i.e., export specialization, trade agreements, and inward migration) embraced by lower-income countries. Hence, the way in which a country manages its tariffs, regulations, and policies to prevent, facilitate, or stimulate international trade, particularly of high-tech and information and communication technologies, reflects in part its openness.

Finally, a critical component of the national innovation capacity is training skilled human capital capable of performing innovative, creative, technical, specialized, and R&D undertakings [72,95,96,97]. Hence, it is essential to establish effective public policies and actions for spending in higher education (i.e., secondary and tertiary education), promoting high-skilled human capital oriented towards innovation, and as a result, economic growth [26,44,93,98,99]. Turpin et al. [99] noted that globalization in finance, higher education, and industrial production has positively triggered the internationalization of technical and scientific human capital. Additionally, Phale et al. [100] found a significant positive correlation between tertiary enrollment, appropriate education spending, government effectiveness, and the presence of technical and scientific journals on economic growth. Moreover, Peercy and Svenson [95] affirmed that increased spending on higher education positively drives innovation and economic development. The investments made in higher education improve a nation’s productive capacity, quality of learning, and competitiveness.

2.2. Innovation Agents

As local and national governments develop public policies to protect IP rights (GS_IP_PROTC), open to international trades (GS_OPEN_INV), and spend on higher education (GS_SPEN_EDU), among other actions, this governmental support provides an engaging environment for innovation agents, enhancing national innovation systems [63,97,101]. As the government sponsors a proper policy framework, the innovation agents leverage such a framework and propose instruments to create a collaborative scenario to foster an innovative ecosystem [46,96,102]. The interaction between universities and industries constitutes one of the most relevant interactions to share ideas, new theories, methods, concepts, models, and resources to partake in innovation and R&D (R&D+i) efforts for mutual advantage (IA_UNI_IND) [44,52,103,104,105,106,107,108,109,110]. In this regard, Giunta et al. [111] investigate the intensity of university-industry interactions on the biopharmaceuticals in Italy and identify their impact on innovation outcomes (scientific articles and patents). They found a positive effect of geographical proximity and prior partnership on the intensity of co-publications. Kurdve et al. [112] found a positive effect of collaborative efforts between SMEs and the university to develop SMEs’ skills and absorptive innovation capacity.

Furthermore, as industries and universities join innovative endeavors, R&D expenditure reflects the economic commitment of both public and private sectors to innovation undertakings. Lee and Kim [113] investigated the influence of several factors, such as R&D expenditure, economic and education freedom, knowledge stocks, among others, on biotechnology innovation. The findings suggested, among other results, that as government R&D sponsored in total R&D expenditure increased, the possibility of biotechnology innovation performance also increased. Nevertheless, an excessive rate of government R&D funding in total R&D expenditure might result in negative innovation effects in the case of developed countries. The results suggested that developing countries require stronger governmental support to promote R&D investment to foster biotechnology innovation.

In this line of reasoning, Tseng et al. [114] pointed to the relevance of scientific knowledge generation and pursuing alliances with universities to improve industries’ innovation outcomes and as a means to gain a competitive advantage. The authors found that the governmental support by placing a variety of programs, financial resources, and policies enhanced the innovation capacity of universities and bridged the gap between industrial applications and academic research. Therefore, as we draw these linkages, the first hypothesis is proposed, as follows:

H1. 

Governmental support will have a positive effect on innovation agents.

2.3. University-Industry R&D Projects

As innovation agents actively partake in R&D+i projects and programs (product and patent development, internships, etc.) to achieve practical innovation outcomes for stakeholders, the national innovation systems also mature, enabling further commitment to undertake riskier innovation endeavors. This commitment required highly skilled human capital to generate and participate in R&D and innovation projects, resulting in successful innovation outcomes [4,26,44,63,93,112,115,116,117]. Hence, it is necessary to hire R&D full-time personnel (UIRD_R&D_PERS) to carry out innovation goals, generate new knowledge (methods, models, theories, mechanisms, artifacts, etc.), and monitor innovation accomplishments daily [46]. UIRD_R&D_PERS are primarily trained in universities, as their function is to transfer knowledge. This highlights the need for instruction and allocation of R&D_PERS to sustain R&D+i operations across industries [80,113] to leverage benchmarking and trade-off opportunities [63,72,118].

By allocating financial resources, the private sectors contribute significantly to disruptive innovation [119]. Thus, R&D spending funded and performed by the private sector and industries (UIRD_GERD_IND) provides a good estimation of businesses’ commitment to innovation [4,26,44]. Previous research has shown a positive effect of UIRD_GERD_IND on companies’ profitability and competitiveness [111,118,120,121,122]. Banelienė [123] found that industry growth in gross value added more strongly boosts GDP growth in highly developed industrialized countries with high GDP per capita than in less-developed ones, especially under GERD performed by businesses. The multiplier effect of such R&D expenditure on economic growth varies by development level. It suggests policymakers in less-developed countries should prioritize improving industry quality through incentives, learning from best practices in advanced industrialized nations, rather than merely increasing the industry’s GDP share.

The maturity of the innovation system is reflected in the allocation of resources that industries, universities, governments, and research centers direct towards innovation undertakings. It is essential to fund innovation endeavors to achieve innovation outcomes and, as a result, improve the innovation capacity of each innovation partner [123,124]. Tseng et al. [114] examined the impact of university-industry collaboration funding on the innovation performance of universities’ technology in Taiwan. To support such collaborations, governmental and private financial support is necessary for resource allocation. They found that collaborative funding was “directly instrumental to universities’ technology innovation.” Therefore, as we draw these linkages, the second and third hypotheses are proposed, as follows:

H2. 

Governmental support will have a positive effect on university-industry R&D projects.

H3. 

The interaction among innovation agents will have a positive effect on university-industry R&D projects.

2.4. Innovation Cluster Development

The state of cluster development refers to the reach and depth of innovation clusters (ICD_CLUST_ST). The innovation cluster’s outline focuses on a sustainable framework that integrates public and private policies, financial sourcing, and investment. It encourages continuous collaborative interactions, fostering innovation, creativity, and R&D activities within a specific industry [4,26,103,111,125,126,127,128,129,130].

This outline is shaped by the openness to trade and investment for R&D+i, and “economic geography” [59,131,132]. The geographical proximity of innovation agents impacts their commitment to innovation undertakings [130]. As knowledge, innovation, and social capital increase, the industrial districts gain competitive advantages, and as a result, higher stages of performance compared to companies outside those districts [133]. Previous research indicated that the interaction among innovation agents and geographic proximity prompt the generation of social capital [134,135]. Hence, the generation of new ideas and knowledge creation and transfer are relevant features of geographic clusters by facilitating technical exchange and knowledge flows [136]. In line with previous studies, Ruiz-Ortega et al. [133] remarked that “the strong competitive dynamic that is generated in the districts of mature industries, as in our case, drives companies to invest in innovation”.

Specialized clusters often give birth to universities’ spin-offs capable of effectively leveraging the use of resources and governmental support [107,130,137,138]. Yoon [36] showed that cooperative efforts between triple helix actors increased across three stages of development. First, industry-government interactions were stronger, shifting later to a more industry-university collaborative orientation, and finally experiencing a more mature triple helix model. Hence, to reach a mature ICD_CLUST_ST, it is required that the active engagement of innovation agents contributes to improving the innovation performance of the innovation system [72,129,139].

Selecting the right cluster counterpart to engage with will potentially impact the achievement and the quality of the pursued innovation outcomes [106,114,138,140]. Therefore, the standing and prominence of innovation agents involved in R&D+i undertakings provide a good measure of their capability and relevance to generate effective innovation outcomes (ICD_PROMIN) [4,25,44,141]. Previous research found a positive impact on innovation outcomes when firms collaborate with prominent and specialized research centers [39,111,112,140,142,143]. Kurdve et al. [112] noticed the importance for Swedish SMEs to engage with the right university to enhance their innovation absorptive capacity. Messeni and Murgia [103] identified the positive impact of collaborations between universities and industries that involve local associates. These collaborations promote knowledge sharing and foreign knowledge integration through these same associates, complementing the resources and knowledge available at the national level. The authors also reported a beneficial moderating effect of university specialization, which can enhance the absorptive capacity of these associates and thereby support an effective application of foreign knowledge.

Finally, Belderbos et al. [140] and Wang and Liu [144] findings point to the importance of a “fit” between the research center specialization, the nature of firms’ research, their strategy, and organizational development to access scientific knowledge to further effective innovation outcomes.

Supported by the theoretical background and conclusions of previous research, the following are the fourth and fifth hypotheses:

H4. 

The interaction among innovation agents will have a positive effect on innovation cluster development.

H5. 

University-Industry R&D projects will have a positive effect on innovation cluster development.

2.5. Knowledge Creation and Knowledge Diffusion

As the innovation system of a nation engages with innovation undertakings, it starts to produce innovation outcomes [4,5,26,60,88,140,145,146]. GII divides the innovation outcomes into two main categories: knowledge and technology outputs and creative outputs [60,147]. This work will focus on two subdivisions of the former: knowledge creation and knowledge diffusion. First, knowledge creation (KC_SCI_PUB) measures the direct generation of new knowledge and technologies. It is related to patents, scientific and technical articles, and utility models. Particularly, technical and scientific patents and articles, including citations of patents, are largely regarded as an indicator of the performance of innovation capacity [26,41,60,148,149,150,151]. Patents (KC_PTNT_PUB/KC_PTNT_APP) are the legal and “exclusive rights granted for an invention” [152]. They comprise the process of creating non-obviousness and novelty in processes and products. The quality of these knowledge creation outcomes depends on the effectiveness of the collaboration between innovation agents and the execution of university-industry R&D projects [107,113,126,127,140,143,153,154].

Furthermore, it is necessary to access financial resources for sponsoring R&D+i processes to generate technical and scientific patents and articles [126,155,156]. Therefore, the innovation cluster development will provide the necessary framework to support the generation of these innovation outcomes [127,157]. Following the same line of argumentation, GII includes knowledge diffusion to measure the ability of an economy to absorb and disseminate knowledge internally and across borders. It is key to ensuring that created knowledge does not remain isolated but prompts innovation across sectors and regions [44,158]. It encompasses the exportation of high-tech (KD_KTCOUT1) and ICTs services exports (KD_KTCOUT2) as part of the measurement of knowledge diffusion [60,92,121,159]. Perez-Trujillo and Lacalle-Calderon [160] found a positive impact of knowledge diffusion on economic growth across countries, and boosts “the technical catch-up process for developing countries” and intensifies their “pattern of economic convergence”. Bartels et al. [161] examined the impact on economic performance as a result of the interaction between four actors from the Ghana national system of innovation, as follows: (1) high- and medium-technology industries; (2) government; (3) arbitrageurs (knowledge brokers, finance capital and venture capital); and (4) knowledge-based institutions. They emphasized the need to empower the relationship between these four actors and promote ICT skills to enhance the overall innovation outcomes. Battaglia and Neirotti [162] studied innovative hi-tech SMEs operating in Italy and confirmed that simultaneous involvement in export and R&D+i activities positively affects the profitability of SMEs, especially when these organizations have broad geographical market reach or collaborate with research centers and universities. However, Autant-Bernard et al. [163] (p. 196) point to the need for designing innovation policies and knowledge diffusion strategies that capture the local features to improve and “exploit public/private, intra/inter-firms, intra/inter-industries and local/global knowledge flows”.

Therefore, our sixth and seventh hypotheses refer to the positive effects of the innovation determinants previously theorized on these innovation outcomes, as follows:

H6. 

Innovation cluster development will have a positive effect on knowledge creation.

H7. 

Innovation cluster development will have a positive effect on knowledge diffusion.

3. Methodology

3.1. Research Model

The research model presents the relationship between governmental support, innovation agents, university-industry R&D projects, and the development of an innovation cluster to achieve knowledge creation and knowledge diffusion as innovation outcomes (Figure 1).

3.2. Instrument and Data

A dataset was designed to gather the data related to the researched constructs. Twenty-one (21) innovation variables were included from the Group of Seven (G7). G7 constitutes an intergovernmental economic and political organization that groups mainly these seven members: the United Kingdom, Italy, Germany, Japan, France, Canada, and the United States. This group was selected due to their heterogeneous culture, practices, industries, and regulatory environment [163,164,165]. This diversity provides a comprehensive scenario for investigating the path toward innovation capacity by capturing different policy frameworks, industries, and innovation systems. It also allows identifying common drivers and barriers to innovation across diverse backgrounds. In contrast, these countries share representative government, liberal democracy, and pluralism as core values [160,166]. This common ground provides a baseline to compare how differences in innovation practices and policies perform under similar governance principles.

Most G7 countries rank among the top twenty most innovative countries (except Italy). Between them, they have established a strong and collaborative network committed to innovation undertakings [167]. This points to their orientation towards innovation, R&D investments, and the well-developed innovation system. Thus, analyzing them enables understanding the factors behind sustained innovation leadership.

Finally, these countries are major IMF advanced economies [168], stressing their significant role in the global economy. Examining the G7 offers insights into the innovation dynamics of global economic leaders, which often set standards and trends for other countries (

Table 1. G7 members GII innovation ranking 2021 and 2024. Source: Own elaboration.

G7 Country	GII 2021 Rank [60]	GII 2024 Rank [44]
Canada	16	14
France	11	12
Germany	10	9
Italy	29	26
Japan	13	13
United Kingdom	4	5
United States of Ameria	3	3

) [52].

Data from 2007 to 2022 were collected and published as a Mendeley dataset [169]. The main sources of information are GII yearly reports, GCI yearly reports, World Bank Database, WIPO Database, and the Organization for Economic Cooperation and Development (OECD), and complemented by Eurostat, the Istituto Nazionale di Statistica from Italy, the Destatis Statisches Bundesamt from Germany, Statistics Canada, and the United States Patent and Trademark Office [170,171,172,173,174,175,176]. The source of each variable is registered in the dataset.

3.3. Procedure

The dataset [169] encompasses the available data of the seven countries published from 2007 to 2022, along with the sources of each innovation determinant included in this study. The original data consisted of 112 records, and after an exclusion process due to a lack of data for certain periods, 69 records were utilized for this analysis. The proposed framework conceptualizes innovation determinants as independent, multidimensional variables rather than manifestations of a single latent construct; thus, a k-means method was applied to cluster the data on a seven-point scale to harmonize the variables [177,178,179]. This method was selected due to the comparative policy focus of this study and the heterogeneous nature of the dataset [180], and the need for interpretable [181,182], robust transformations across multiple innovation determinants [183]. This technique allowed us to preserve the variable-specific structure while enabling standardized analysis across dimensions.

3.4. Measurement of Constructs

From Section 2, five constructs were identified to trace a pathway towards innovation capacity: government support, innovation agents, university-industry innovation projects, innovation cluster development, knowledge creation, and knowledge diffusion. Also from the previous section,

Table 2. List of Items of each construct. Source: Own elaboration.

Constructs	Item	Label	Description	References
Governmental Support	Intellectual property protection	GS_IP_PROTC	World Economic Forum, Executive Opinion Survey. Response to the survey question “In your country, to what extent is intellectual property protected?” [1 = not at all; 7 = to a great extent]|2017–2018 weighted average or most recent period available.	[26,44,108]
	Imports of goods and services	GS_OPEN_INV	World Trade Organization (WTO). Imports of goods and services expressed as a percentage of GDP.	[4,26,108,175]
	Expenditure on education	GS_SPEN_EDU	Government expenditure on education (% of GDP).	[26,60]
Innovation Agents	Gross expenditure on R&D	IA_GERD_GEN	GERD refers to the “total domestic intramural expenditure on R&D during a given period as a percentage of GDP”.	[26,60,173]
	University–industry R&D collaboration	IA_UNI_IND	World Economic Forum, Executive Opinion Survey. Average answer to the survey question: In your country, to what extent do businesses and universities collaborate on research and development (R&D)? [1 = not at all; 7 = to a great extent].	[4,60,108]
U-I R&D projects	GERD performed by business	UIRD_GERD_IND	GERD performed by business enterprise as a percentage of total gross expenditure on R&D.	[60]
	Researchers FTE/mn pop	UIRD_R&D_PERS	Full-time research personnel (per million population).	[26,60,173]
Innovation cluster development	Research institutions prominence	ICD_PROMIN	World Economic Forum, Executive Opinion Survey. In your country, how do you assess the quality of scientific research institutions? [1 = extremely poor—among the worst in the world; 7 = extremely good—among the best in the world]|2016–2017 weighted average.	[4,108,184]
	State of cluster development and depth	ICD_CLUST_ST	World Economic Forum, Executive Opinion Survey. Average answer to the survey question: In your country, how widespread are well-developed and deep clusters (geographic concentrations of firms, suppliers, producers of related products and services, and specialized institutions in a particular field)? [1 = nonexistent; 7 = widespread in many fields].	[60,108]
Knowledge creation	Scientific Publications	KC_SCI_PUB	Number of scientific and technical journal articles (per billion PPP USD GDP).	[60,185,186]
	Patent applications	KC_PTNT_PUB	Intellectual property right. Patent Total Resident.	[172,186]
	Patents applications by origin	KC_PTNT_APP	Total count by applicant’s origin (equivalent count).	[172,186,187]
Knowledge diffusion	High-tech exports	KD_KTCOUT1	High-tech exports (percentage of total trade).	[60,185]
	ICT services exports	KD_KTCOUT2	Telecommunications, computers, and information services export (percentage of total trade).	[60,185]

presents the list of items (innovation determinants) related to each construct and the reference to previous studies to support them.

4. Empirical Results

IBM SPSS Statistics 26.0 was the tool to process the database. The final sample consists of 69 records, given the incomplete and invalid responses. The analysis included a descriptive statistical analysis, a correlation analysis of the variables, and a regression analysis [171].

4.1. Descriptive Statistics

Table 3. Descriptive statistics. Source: Own elaboration.

Variables	Mean	Std. Dev.	Min	Lower Quartile (Q1)	Median	Upper Quartile (Q3)	Max
Governmental Support (GS)	4.478	1.212	2.333	3.167	5.000	5.500	6.000
Innovation Agents (IA)	4.507	1.670	1.000	3.500	4.500	6.000	6.500
U-I R&D projects (UIRD)	4.928	1.709	1.000	4.500	5.000	6.000	7.000
Innovation cluster development (ICD)	5.174	1.074	3.000	4.500	5.500	6.000	7.000
Knowledge creation (KC)	2.957	0.941	1.667	2.333	2.667	3.667	5.333
Knowledge diffusion (KD)	3.645	1.047	2.000	3.000	3.500	4.000	6.000

Note: N = 69.

shows the statistical characteristics of this work. On the seven-point scale, the composite mean score of ICD presented the highest value (µ = 5.174), showing the strength of innovation cluster development among G7 members and thus the relevance of geographical proximity and specialization among the innovation agents participating in their national innovation systems. The ICD SD value is 1.074, which reflects a relative concentration of the observations around the mean and indicates a homogeneous level of ICD maturity among G7 countries.

Furthermore, the UIRD composite score is 4.928, reflecting the commitment of universities and industry to innovation by hiring full-time skilled personnel dedicated to developing R&D+i projects while effectively managing financial resources. However, UIRD has the highest SD value (1.709). The values are widely spread, pointing to the heterogeneity and prominent differences of the countries in the collaboration between universities and industries to undertake R&D+i projects.

The composite score for IA is 4.507, revealing the relevance of the collaborative efforts and expenditure between academia and industry on R&D to achieve innovation outcomes. The SD value related to IA is 1.670, displaying high dispersion, with values ranging from 1 to 6.5 on how the group of countries undertakes the collaboration between university and industries as innovation agents within their national innovation system. This suggests that while in some countries, their environments have highly collaborative innovation agents, others may struggle to integrate effectively.

GS registered a composite score of 4.478, indicating a balanced policy framework promoted by the governmental infrastructure to protect intellectual property, provide an open environment to international trade, and invest in education to strengthen the national innovation system. The SD value of this construct (σ = 1.212) reflects moderate dispersion in GS between the seven countries, implying some variation among their policy frameworks and openness levels.

Regarding the innovation outcomes constructs, the composite scores for KD (µ = 3.645) and the SD value (σ = 1.047) are moderate, suggesting that although KD is not among the top strengths, countries are open to sharing their innovation developments abroad. Finally, the KC composite score is the lowest (µ = 2.957), implying a relatively weak level of KC in the observed countries. Also, this construct presented the lowest SD value (σ = 0.941), indicating a strong homogeneity of KC level among G7 countries. These results picture the shared difficulty in achieving innovative outputs by these countries.

4.2. Correlation Analysis

Table 4. Correlation matrix. Source: Own elaboration.

	GS	IA	UIRD	ICD	KC	KD
GS	1.000
IA	0.237 *	1.000
UIRD	0.281 **	0.873 ***	1.000
ICD	−0.003	0.721 ***	0.496 ***	1.000
KC	−0.262 **	0.658 ***	0.734 ***	0.420 ***	1.000
KD	0.543 ***	−0.047	−0.006	−0.333 ***	−0.479 ***	1.000

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels (two-tailed), respectively.

presents the correlations between all constructs. The results regarding GS correlations indicate a positive impact on IA (0.237 *), collaborative R&D projects (0.281 **), and strongly influence KD (0.543 ***). In contrast, there is a negative correlation with KC (−0.262 **).

Moreover, IA results show a strong positive correlation with UIRD (0.873 ***), ICD (0.721 ***), and KC (0.658 ***), implying the central role played by IA in fostering collaborative R&D projects, cluster development, and knowledge creation. UIRD results present a strong positive relationship with IA (0.873 ***) and links with KC (0.734 ***) and ICD (0.496 ***), pointing towards its significant boost to KC and the presence of linkages between collaboration and cluster development. ICD presents a moderate correlation with UIRC and KC, implying that cluster development benefits from active innovation agents and collaborative projects. Nevertheless, there is an indication of a negative correlation with KD (−0.333 ***), suggesting that clusters, while good for internal knowledge creation, may not contribute directly to knowledge diffusion in this context.

Regarding innovation outcomes, KC presents strong positive correlations with IA (0.658 ***), UIRD (0.734 ***), and moderate with ICD (0.420 ***), reinforcing the role of collaboration and clusters in fostering knowledge creation. On the other hand, there is a negative correlation with KD (−0.479 ***), suggesting some differences in innovation processes, where in this context, the countries may prioritize KC over KD. Finally, the results point to a strong positive correlation between KD and GS (0.543 ***), indicating the positive impact of governmental policies to enhance the dissemination of knowledge. In contrast, as pointed out previously, there is a negative correlation with ICD and KC, implying that diffusion appears to occur more from policy and less from clusters or knowledge creators.

From the correlation matrix, it is possible to identify potential multicollinearity between IA and UIRD (r = 0.873 ***), as the value is above the 0.8 threshold. However, in the regression analysis, it is shown that the multicollinearity problem does not prevail (see

Table 5. Relationship between variables. Source: Own elaboration.

Regression	Prop Effect	Adj. R2	F	Constant	β	VIF	Test Result
GS → IA	+	0.042	3.979 ***	3.046 ***	0.326 **	1.000	H1 supported
				(4.015)	(1.995)
GS → UIRD	+	0.065	5.737 ***	3.154 ***	0.396 **	1.000	H2 supported
				(4.113)	(2.395)
IA → UIRD	+	0.759	215.210 ***	0.900 ***	0.893 ***	1.000	H3 supported
				(3.078)	(14.670)
IA → ICD	+	0.512	72.424 ***	3.086 ***	0.463 ***	1.000	H4 supported
				(11.801)	(8.510)
UIRD → ICD	+	0.234	21.831 ***	3.639 ***	0.311 ***	1.000	H5 supported
				(10.475)	(4.672)
ICD → KC	+	0.164	14.348 ***	1.053 ***	0.368 ***	1.000	H6 supported
				(2.051)	(3.788)
ICD → KD	+	0.098	8.377 ***	5.327 ***	−0.325 ***	1.000	H7 not supported
				(8.979)	(−2.894)

Note: Beta corresponds to unstandardized coefficients. Variance inflation factor (VIF). Numbers inside the parentheses are t-statistics. + indicates an expected positive relationship between variables. ***, and ** indicate statistical significance at the 1%, and 5% levels (two-tailed), respectively.

, Variance Inflation Factor (VIF) results). Another potential multicollinearity risk is ICD and IA due to their close but below-the-threshold values (0.721 ***), avoiding major concern. Hence, most relationships are acceptable and theoretically meaningful.

4.3. Regression Analysis

Table 5 presents the findings of the individual linear regressions to test the effect of governmental support, innovation agents, R&D+i projectsI confirm and innovation cluster development on knowledge-and-technology innovation outputs. The VIF was calculated for all independent variables included in each regression model to ensure the multicollinearity problem did not prevail (particularly between IA and UIRD) [188]. All computed VIF values were equal to 1.00, which is well below the conventional thresholds of 5 or 10 [189,190]. The highest adjusted R-Square is 0.759 (H3 result), and the lowest value is 0.042 (H6 result).

Six hypotheses of the study are supported. First, H1 and H2 are supported as the results showed a positive but modest influence of governmental support on innovation agents (β = 0.326, p < 0.05) and university-industry R&D collaboration (β = 0.396, p < 0.05), with relatively low explanatory power (adjusted R² = 0.042 and 0.065, respectively). These findings suggest that while governmental efforts are necessary, as they provide a suitable environment for promoting intellectual property protection, and openness towards international trade and FDI to foster the engagement of innovation agents such as universities, industries, and R&D centers with R&D+i undertakings [63,82,85,101,102,105,191]. However, similar to previous studies, these results point to a fine line to balance the proper level of government support before it negatively impacts innovation outcomes [113].

Moving forward, H3 is supported where the relationship between innovation agents and university and industry R&D projects was strong (β = 0.893, p < 0.001), accounting for nearly 76% of the variance (adjusted R² = 0.759). This emphasizes the pivotal role of university-industry as collaborative agents fostering R&D projects by allocating highly skilled human capital [63,112,115,131,192] and financial resources [119,123,193]. Moreover, innovation agents have a positive effect on cluster development, supporting H4 (β = 0.463, adjusted R² = 0.512). As the government provides a balanced framework to support innovation efforts, innovation agents leverage the available policies and resources to deepen the innovation cluster [107,130,137,138]. This reinforces the central role of innovation agents as the driving force in the model, by shaping national and regional innovation ecosystems [80,114,192,194].

Regarding university-industry R&D projects, these also contributed significantly to cluster development (β = 0.311, p < 0.001), though with a lower explanatory power (adjusted R² = 0.234), indicating that R&D collaboration is a supporting, yet not dominant, component in the development of clusters. The subsequent path from cluster development to knowledge creation was significant (β = 0.368, p < 0.001) (H6 supported), suggesting that innovation clusters foster environments conducive to generating new knowledge [149,154,195,196], though with moderate explanatory capacity (adjusted R² = 0.164).

Finally, H7 was not supported. Contrary to expectations, innovation cluster development showed a significant negative effect on knowledge diffusion (β = –0.325, p < 0.001). This counterintuitive result may point to a tendency for clusters to operate as closed systems, where knowledge is developed but not widely shared outside the immediate network.

5. Discussion

Several countries have undertaken the development of innovation capacity as a strategic approach for economic growth [197,198].

This work examined the relationship between governmental support that frames R&D+i policies and activities, innovation agents that leverage available resources to develop R&D projects, and the state of development of innovation clusters. The linkages between these factors have an impact on knowledge creation (e.g., patents, scientific articles) and knowledge diffusion (e.g., high-tech exports, ICT services exports) as innovation outputs.

This research examined the role of governmental support as a starting point to provide a basic framework for supporting innovation agents’ commitment to innovation undertakings. In line with previous research, the results exhibit the need for proper governmental support by providing IP protection mechanisms [78,80,81,83,85], promoting openness to international trade [42,86,87,88] and expenditure on higher education to trigger [63,101,106,199]. However, the findings also suggest that once governmental support reaches a threshold, its impact becomes counterproductive to what was expected [113]. These results have practical implications for both developed and developing countries. Developed countries, such as G7 members, have mature institutions and policy frameworks providing an effective environment to trigger innovation agents. In contrast, developing countries with limited government capacity, inconsistent policy implementation, and institutional weaknesses often face a lack of effectiveness of governmental support. Studies show that developing countries frequently confront weak IP enforcement, poor governance, and limited public funding for education and R&D [200,201,202,203]. The “threshold effect” observed in G7 countries may not yet apply in many developing countries, where support is often insufficient or misaligned with local needs [85,200].

Furthermore, the results identified the critical role of universities and industries as innovation agents driving collaborative R&D projects. These innovation agents allocate human talent and financial resources as they engage with collaborative R&D projects, supporting previous findings [114,123,124]. These interactions, the allocation of resources, and effective collaboration with R&D projects showed a positive impact on the development and depth of the innovation cluster of the seven countries, in line with former studies [72,129,139]. This suggests that while governmental backing lays the groundwork, its effectiveness appears to depend largely on how it stimulates and strengthens main innovation agents to foster innovation clusters [204,205]. For developed countries, this highlights the maturity of universities and industries as established innovation agents that collaborate on effective R&D projects, fostering specialized innovation clusters with relevant knowledge creation [163,206]. However, developing countries are often limited by weak university-industry collaboration, low technology and science intensity, and skilled human capital [21,72,207]. As a result, innovation actors are less committed to undertaking R&D projects, which are mostly informal and less frequent [208]. This further hinders cluster development and the depth of innovation outcomes. Local innovation often takes place outside formal R&D structures and tends to be localized and incremental [201,209]. Following the implications of these results, knowledge flows and international collaboration are essential for strengthening the national innovation capacity, especially for the countries seeking to catch up or leapfrog in technological development [163,210]. This collaborative approach enables countries to share data, infrastructure, and financial resources, reducing duplicated efforts while increasing efficiency [211]. This also leads to the configuration of international, skilled, and knowledge-based teams. Furthermore, for countries with limited R&D structure, openness to global knowledge may allow them to avoid transitional stages of technological development, adopting advanced solutions directly [212]. Developing countries can benefit from catch-up mechanisms by absorbing knowledge spillovers from developed countries’ firms and internationalizing their own R&D undertakings, gradually narrowing the technological gap [203,213,214].

Regarding the achievement of innovation outcomes, innovation cluster development among the seven countries under study showed a positive impact of innovation on achieving knowledge creation while posing a negative impact on knowledge diffusion. These findings may reflect a trade-off between depth and breadth of knowledge processes—clusters that focus intensively on specialized knowledge creation may inadvertently limit broader dissemination [171]. These countries have largely focused their efforts on developing knowledge, while Asian countries have a strong focus on manufacturing and exporting ICTs and high-tech production [9,171,215,216]. This warrants further qualitative or structural exploration. In comparison, innovation clusters in developing countries are often fragmented and less mature, limiting both knowledge creation and diffusion. Moreover, these countries rely heavily on technology transfer and adaptation from abroad rather than purely local knowledge creation [201,217]. The diffusion of knowledge is often constrained by poor infrastructure, limited connectivity to global innovation networks, and low absorptive capacity [170,200,217].

Turning to the unexpected negative correlation of knowledge diffusion with innovation cluster development, it has important theoretical and practical implications grounded in the multifaceted nature of innovation national systems. On the one hand, innovation clusters focus on knowledge creation, producing specialized and cutting-edge knowledge as a result of the geographical proximity of innovation agents. On the other hand, knowledge diffusion aims to transfer and spread knowledge beyond cluster boundaries by commercializing, adopting, and spilling over innovations, resulting in broader social and economic benefits [163,218]. Both innovation outcomes are complementary; thus, assessing them separately acknowledges that creating knowledge does not automatically assure its widespread dissemination. Moreover, the potential trade-off between these innovation outcomes may occur due to knowledge protection strategies. The tacit nature of knowledge requires close, trust, and committed interactions to be transferred effectively [219]. Furthermore, clusters may prioritize knowledge sharing and internal collaboration, enhancing creation but hindering outward diffusion, particularly if the incentives are weak. This emphasizes the need for a more in-depth understanding of the knowledge transfer mechanisms and factors that simplify or limit diffusion. By observing these two constructs separately, policymakers and cluster managers can identify gaps in diffusion and design interventions (e.g., mobility, open innovation platforms, or promoting inter-clustering networks) to improve knowledge flows. It also provides a better perspective of the dynamics across the innovation ecosystem, enabling the design of balanced cluster policies to foster creation while defining strategies for the dissemination and adoption of innovations. This will lead to maximizing the economic impact and social welfare of innovation outcomes.

6. Conclusions and Limitations

This investigation explored a pathway to sustain the development of innovation as a national strategic capacity to foster competitive advantage. For a nation, it is critical to identify a configuration of its innovation national framework where policies, innovation agents, resources, and skills are effectively combined to achieve successful innovation outcomes. However, assessing innovation is a complex task due to the multi-faceted nature of this capacity. Hence, G7 country members were selected. From an extensive theoretical background and a regression analysis with data collected from 2008 to 2018, a practical pathway towards innovation was identified. These countries rank among the thirty most innovative countries by GII 2024, providing a comparative analysis between heterogeneous cultures, practices, and industries while sharing political and institutional values.

Defining the pathway towards innovation requires a deep understanding of regional and local innovation policies, leading to supporting innovation agents to generate knowledge. The findings emphasize the impact of balanced governmental support, strong innovation agents, and cluster development to achieve sophisticated innovation outcomes. From the discussion, we can conclude that developed countries with strong innovation systems and high R&D intensity can sustain complex innovation pathways. Meanwhile, many developing countries face structural difficulties, including limited resources and infrastructure, which restrict their innovation capacity and outputs. Hence, the innovation roadmap for these countries may prioritize establishing basic institutional and human capital capacities, policies, and institutional values before expecting higher levels of cluster sophistication. They should address critical gaps and provide the mechanism to enhance technology transfer and absorptive capacity, leveraging foreign technologies effectively. Also, it must foster public and incremental innovation addressing local socioeconomic challenges. All this requires putting in place tailored policies to frame country-specific contexts.

As a theoretical contribution, this research provides exploratory empirical evidence to ground theory regarding the development of innovation, as a national capacity, in a sustainable fashion to enhance the competitive advantage of a nation. Taken together, these results emphasize the importance of intermediary innovation actors such as universities and industries in driving innovation and collaborative endeavors. It is also important to keep in mind that there is a threshold point where the excess of government support becomes ineffective.

This work offers practical insights into the advantages that highly innovative and developed countries gain by establishing a comprehensive framework to enhance their innovation capabilities. This framework requires a balanced approach that includes effective government support through policies designed to promote the development of cohesive innovation clusters, leading to sophisticated innovation outcomes. Practitioners and policymakers may use these findings to create a roadmap aimed at strengthening similar linkages in developing countries. Additionally, the differing impacts of innovation clusters on knowledge creation and diffusion highlight a complex dynamic that necessitates deliberate strategies to ensure both deep innovation and broad outreach are successfully achieved.

This study is exploratory in nature with some limitations. The proposed construct structure is grounded in theory and prior research. Each variable was chosen for its relevance to the theoretical understanding of innovation ecosystems, based on consultations with innovation policy and prior use in applied research. Governmental support, innovation agents, and university-industry R&D projects have all been identified as key components of regional innovation systems [220]. We limited the sample size to seven countries that have shown a consistent approach towards innovation. However, this confines the applicability of some of the findings to developing countries as previously discussed. Moreover, this highlights the need for further research to assess the proposed pathway, considering countries at different stages of innovation.

The process of data collection was limited to the available reports published yearly that tracked the same measure of each construct included. Furthermore, some values were properly reported for certain years. We employed multiple linear regression as a robust and transparent method to explore the direct relationships between the constructs. This approach is suitable for handling the sample size per country-year and facilitating comparison and interpretability. Further comparative studies could employ more advanced techniques, such as structural equation modeling (SEM) or path analysis, to examine hierarchical relationships between constructs and larger datasets. Also, artificial intelligence tools could be used to deepen this understanding from several perspectives.

Future studies should investigate the causal mechanisms behind these relationships, assessing potential mediating or moderating factors. It is also necessary to expand the study of this field by including specific innovation factors that influence economic indicators such as GDP, inflation, interest rates, and unemployment. Another emerging field of study related to the development of innovation capacity is the analysis of digital indices such as the artificial intelligence index and the digital sustainability framework. Therefore, future research could examine the impact of these variables on the sustainability of national innovation capacity.