Determinants of Patent Activity and the Formation of Comparative Advantages in Selected European Economies

Abstract

This paper examines the relationship between innovation performance and international export competitiveness in four EU countries—Germany, Italy, Czechia, and Slovakia—during the period 2015–2024. The primary objective is to identify the relationship between the number of patent applications to the European Patent Office (EPO) and two key R&D input indicators: R&D expenditure per capita and the number of researchers and engineers per million inhabitants. Simultaneously, the study utilizes the Revealed Comparative Advantage (RCA) index to evaluate export specialization in medium-to-high innovation-intensive commodity groups. Although there are numerous studies on innovation, patents, or the significance of research and development, only rarely are these indicators linked to the competitiveness of countries according to comparative advantages in individual sectors. The results of the correlation analysis reveal significant national disparities: while a strong dependency was confirmed in Italy and Slovakia, the findings for Germany show a negative correlation, suggesting that German patenting activity is driven by factors beyond the examined R&D inputs. Panel regression also points out that simple correlation may not be able to clearly capture this relationship, as it may manifest itself with a time lag. From an absolute perspective, Germany maintains a leading position in all indicators, yet Italy demonstrates higher patent efficiency compared to Czechia despite having fewer researchers. The RCA analysis further highlights that while Germany and Italy maintain comparative advantages in high-innovation sectors, Czechia and Slovakia predominantly specialize in medium-innovation-intensive industries.

1. Introduction

A primary objective of economic policy in developed nations is to ensure economic competitiveness, which is fundamentally driven by innovation performance. In the contemporary global landscape, innovation has emerged as a critical element of competitive advantage (Prokop & Stejskal, 2017). Innovation is the driving force of each economy, competitiveness of the economy and a main essential component of the knowledge economy (Kučera & Fiľa, 2022). In the period of growing globalization and digitalization, innovation is becoming a more important factor in determining the success of business activity and countries (Belanová, 2021).

R&D serves as a fundamental prerequisite for the creation and advancement of innovations, representing a hallmark of the knowledge-based economy and underpinning its innovative capacity. R&D expenditures are considered to be efficiently utilized only when successfully transformed into innovations that enhance overall economic competitiveness.

R&D inputs are typically assessed based on financial allocations to research activities as well as the quality and composition of human capital. R&D expenditures have been proven to be a key determinant of innovation activities in all developed countries. These are primarily private sources of in-house research, intramural government expenditure on R&D, research and development spending by universities and the public sector (e.g., public laboratories and research institutes) (Prokop et al., 2019).

Both financial investments and specialized human resources in this field constitute essential preconditions for improving a nation’s innovation performance. Innovation performance can be measured using various metrics; academic literature widely identifies the number of patent applications as a key proxy for measuring innovative output.

This study aims to evaluate the relationship between patent applications submitted to the European Patent Office and two key variables—R&D expenditure and the number of scientists and engineers—across four European countries: Czechia, Germany, Italy, and Slovakia. Concurrently, the study employs the Revealed Comparative Advantage index to identify the comparative advantages in the export of selected commodity groups originating from medium-to-high innovation-intensive industries.

Although there are many studies focusing on innovation performance, patent activity or R&D investment (Chu, 2021; Leogrande et al., 2022; Nie et al., 2023; Yang et al., 2024; Jose et al., 2022; Levratto & Quignon, 2020), only a few of them link these indicators with the competitiveness of countries according to comparative advantages in individual sectors (Guan & Chen, 2012). This article therefore deals with linking innovation inputs and outputs with the export specialization of countries, considering the different technological complexity of sectors. The main objective of this study is to identify the relationship between the number of patent applications to the EPO per million inhabitants and the amount of research and development expenditure per capita, or the number of scientists and engineers per million inhabitants in four European countries (the Czech Republic, Germany, Italy and Slovakia), and at the same time, through RCA, to identify their comparative advantages in the export of selected commodity groups of goods originating from sectors with medium-to-high innovation requirements.

The article is structured into several parts. In the first part, the theoretical background and basic concepts related to the research issue are analyzed. This is followed by a section dedicated to the research methodology, in which we describe the data used and analytical procedures. In the third part, the research results are presented and in the Discussion section, findings with existing studies in the given area are compared. The conclusion contains a summary of the most important results and suggests recommendations for further research, based on the identified limitations of this study.

2. Theoretical Framework

Continuous changes in the global economic landscape, intensifying competition, and rapid technological progress necessitate ongoing impulses in the form of innovation. Innovation is a managed process of generating, transferring, and implementing ideas into practical applications, leading to enhanced quality in products, services, and processes. It is the result of realizing creativity, new ideas, and original knowledge (invention). Fundamentally, it involves applying progressive ideas or methods in practice to generate financial returns. Innovation serves as the driving force of every economy, its competitiveness, and is a fundamental component of the knowledge economy, which is built on high value-added production and the support of research and development (Kučera & Fiľa, 2022). Several authors emphasize the importance of innovation for increasing competitiveness; Prokop and Stejskal (2017) argue that innovation plays a vital role in most national policies, as it clearly leads to increased welfare and the competitiveness of individual economic entities.

The role of innovation is to translate research results into new and improved services and products to maintain competitiveness in the global market and enhance citizens’ quality of life (Kordoš & Krajňáková, 2018). Innovation is closely linked to science and research, representing the practical realization of an idea or concept. The sequence of steps—from the creation of an invention through innovation development to market launch—is termed the innovation process. Its final output is innovation performance, which is evaluated at the firm, national, and regional levels.

From a macro-theoretical perspective, the necessity of innovation is rooted in endogenous growth theory, which posits that long-term economic expansion is driven by deliberate investments in human capital and knowledge creation (Romer, 1990). Within this framework, innovation is not an exogenous shock but an internal output of national innovation systems, where synergy between institutions, universities, and firms determines a country’s technological trajectory. This process is inextricably linked to Schumpeterian competition, characterized by “creative destruction,” where firms must innovate to secure temporary monopoly rents, thereby pushing the entire economy toward higher technological frontiers and bridging the gap between leading and lagging nations.

Research and development form the basis for the emergence and advancement of innovations. As stated by Duľová Spišáková et al. (2021), R&D is a key mechanism for economic revitalization, entrepreneurial activity, and success amidst fierce competition in both domestic and international markets. Odei et al. (2025) note that R&D serves as a catalyst for innovation, technological progress, and, subsequently, sustainable development. A prerequisite for conducting high-quality R&D is the availability of sufficient financial resources; however, it is not only the absolute volume of R&D investment that matters but also its underlying structure.

The impact of R&D on innovation performance, economic growth, and competitiveness is the subject of extensive scientific research. Szarowska and Žurkova (2017) examined public R&D expenditure and its link to economic growth, specifically in the Czech Republic, Denmark, and Slovakia. Their results confirmed a positive long-term relationship between gross domestic expenditure on R&D and economic growth in Denmark and Slovakia.

Teng and Yi (2017) investigated the influence of ownership types on R&D intensity and innovation performance of Chinese firms. They found that central government-owned firms are key drivers of corporate R&D activities, whereas local government, private, and foreign ownership correlated negatively with both R&D intensity and innovation performance. Their research thus suggested that concentrated state ownership correlates positively with a firm’s R&D intensity and innovation output.

Research into the relationship between innovation performance and R&D expenditure spans firm, regional, and macroeconomic levels. Aarstad and Kvitastein (2019) explored how the link between R&D investment and innovation performance is moderated by a firm’s location within a specialized or diversified regional industrial structure in Norway. They found that innovation performance resulting from corporate R&D investment is stronger in regions with a specialized structure compared to those with a diversified structure. A specialized regional industrial structure embodies a specialized workforce with similar and coherent knowledge, thereby creating cognitive proximity and a shared understanding of the R&D vision. Furthermore, Blažek and Kadlec (2019) examined the relationships between knowledge bases, R&D structure, and innovation performance in European regions. They found that developed regions are often characterized by the lowest share of synthetic knowledge bases and either the dominance of private R&D or a relatively balanced structure between private and public R&D, whereas the opposite holds for lagging regions.

Furthermore, in the context of innovation performance, it is essential to incorporate the concept of intrapreneurship (corporate entrepreneurship), which serves as a critical mediating mechanism between R&D resources and resulting competitiveness. Technological development and R&D investments represent the fundamental building blocks of intrapreneurship, as they provide employees with the necessary knowledge base and tools to identify and exploit new market opportunities.

As scholarly literature suggests, a high level of intrapreneurial activity enables firms to more effectively commercialize R&D outputs and transform patents into viable market products. Intrapreneurship is thus not merely a byproduct of available technology, but a dynamic capability that reactivates untapped knowledge within an organization. In the context of international competitiveness, nations with environments conducive to intrapreneurship are better positioned to rapidly transform technological inputs into sectoral export specialization (RCA).

While R&D provides technical solutions, intrapreneurship ensures their strategic implementation. Antoncic and Hisrich (2001) define intrapreneurship as a process involving not only product innovation but also organizational self-renewal, which is crucial for sustaining long-term international competitiveness. This integrative perspective is further supported by Neessen et al. (2019), who state that individual employee characteristics, combined with a supportive R&D environment, are key determinants in the successful transformation of technological knowledge into commercial outputs.

Raghupathi and Raghupathi (2019) examine innovation at the country level using OECD data on R&D, patents, and exports. They identified significant differences between developed (high-income) and developing (middle-income) countries regarding sectoral R&D expenditures. While R&D spending in developing countries stems predominantly from the government sector, in developed countries, it is primarily driven by the business sector. As a country develops, the share of government R&D expenditure tends to decrease, with businesses and educational institutions playing a more prominent role in bridging this gap.

Various perspectives exist for assessing innovation performance, both from institutional frameworks and individual researchers. The European Commission evaluates innovation through the European Innovation Scoreboard, which utilizes the Summary Innovation Index. Based on these data, Janošková and Kráľ (2019) evaluated the innovation performance of V4 countries between 2010 and 2016. Similarly, Belanová (2021) used this index to analyze Slovakia’s innovation performance compared to the European Union average.

As emphasized by Barrichello et al. (2020), the innovation process is a multifaceted phenomenon influenced by a wide range of determinants. These include a nation’s overall innovation capacity, the quality of its scientific research institutions, and the level of business R&D investment. Additionally, they highlight the importance of university–industry collaboration, government procurement of advanced technological products, and the availability of specialized human capital, particularly scientists and engineers. Within this framework, Barrichello et al. (2020) argue that the rate of innovation is most effectively quantified through two primary indicators: R&D expenditures and patent application volume, which serve as standard metrics for assessing innovative progress.

In many studies, the number of patent applications serves as a frequently used indicator of innovation performance. Chen et al. (2025) emphasize that high-quality patents are key intangible assets that grant inventors exclusive rights, attract investment, and ensure market exclusivity, which is essential for firms to secure a technological competitive advantage. Chu (2021) focused on the relationship between patent policy, innovation, and economic growth, concluding that the patent system is an important policy tool for stimulating innovation given its significance for technological progress. They also noted that the positive correlation between patent rights and innovation is more pronounced in developed countries.

Yildiz and Görkey (2024) consider R&D activities crucial for securing economic growth, although they serve primarily as inputs for innovative activities, whereas patents are seen as the output of such efforts—the final version of the entire R&D endeavor. According to Y. Liu et al. (2023), the number of patents is the most direct indicator for measuring the production of new knowledge and technologies. Diebolt and Hippe (2019) utilized patent counts to measure innovation levels, and similarly, Levratto and Quignon (2020) state that patents serve as a proxy for innovation results. Patent activities are highly reliable indicators of the economic value of knowledge and firm performance. In his research, Klas (2010) identified a dependency between the innovation growth (measured by patents) and the growth of economic performance.

Zhang et al. (2020) note that new products and patents are often used to measure innovation performance. While sales revenue from new products reflects the value created in the production process, the number of patents reflects the inventive or innovative performance of enterprises in terms of new technologies, processes, and products. Dziallas and Blind (2019) categorize patent counts and R&D budgets as indirect indicators of innovation assessment, whereas direct indicators include the number of new product ideas and the percentage of ideas with commercialization potential. Yang et al. (2024) identify patents and publications as the two primary outputs of scientific and technological innovation. They argue that patents motivate inventors to invest in R&D for sustainable technologies by providing exclusive rights for a limited period.

According to Rahko (2016), patent inventor data are used to track the geographical distribution of corporate R&D activities. Information on patent applications, citations, and technological fields is used to measure the output, quality, and diversity of innovation. Leogrande et al. (2022) hypothesized that patent rights play a role in supporting technological innovation and economic growth; however, their analysis revealed that European business systems—excluding the highest-income countries—are not effectively generating high levels of patented technological innovation. Consequently, they emphasize the need for increased incentives for companies to boost patent production.

Research by Huňady and Orviská (2014) shows that R&D expenditures in EU Member States correlate positively with both the number of R&D personnel and the number of patents. This suggests that financial inputs into R&D largely lead to higher employment in this sector, which subsequently results in higher patent output.

Despite the widespread use of patents as an indicator of innovation performance, several authors express academic reservations. Martínez-Noya and García-Canal (2021) caution that although patents are considered a valid indicator of technological competence in high-tech industries, their use has inherent limitations. They point out that many patents are never commercialized, and some innovations are either not patentable or not worth the cost of patenting. Similarly, Seoh and Im (2020) warn that a patent is merely an idea that may or may not be incorporated into a final product; thus, the number of patents may not fully reflect R&D performance. They suggest that patents are intermediate products rather than final outputs of R&D activities—representing innovative ideas rather than tangible results. Despite these reservations, the use of patents as a measure of innovation performance remains widespread, and their link to R&D expenditure and personnel is indisputable.

Innovation is considered a significant factor in economic growth and productivity, which subsequently impacts the growth of international competitiveness. According to Porter (as cited in Tadevosyan, 2023), international competitiveness depends on a country’s industry’s ability to innovate and upgrade. As Vu et al. (2025) state, innovation capabilities at macro and micro levels significantly increase a state’s export performance, with innovations providing firms with a competitive advantage in international markets. This reality points to a transition from traditional comparative advantages, such as low labor costs, to advantages based on innovation and innovation performance. Innovations thus transform traditional comparative advantages into dynamic competitive advantages. Higher innovation performance subsequently increases a country’s ability to export goods with higher levels of sophistication and quality. To measure international competitiveness, scholarly literature frequently uses the Revealed Comparative Advantage (RCA) Index, which reveals a country’s specialization in a specific industry or product. This index, popularized by Balassa (1965), is rooted in the assumption that trade patterns reflect not only relative factor abundance but also differences in technological levels. It is considered one of the most widely used indicators of international competitiveness (Hasan et al., 2024), despite the development of other variations intended to eliminate its methodological shortcomings. In alignment with the Technology Gap Theory of trade and the concept of structural transformation, a shift in RCA towards high-technology intensity sectors (as classified by the OECD) indicates an economy’s capacity to overcome technological barriers and compete through knowledge rather than merely price.

As previously mentioned, numerous studies focus on evaluating innovation performance (Aarstad & Kvitastein, 2019; Atsu & Adams, 2023; Blažek & Kadlec, 2019; Gruzina et al., 2021), patent activity (Chen et al., 2025; Chu, 2021; Y. Liu et al., 2023), or R&D investments (Diakodimitriou et al., 2025), Most of these monitor these areas separately and do not link them with an analysis of international competitiveness, although research focused on the link between the impact of innovation performance and competitiveness on export competitiveness is beginning to emerge (e.g., Salamaga, 2020; Tadevosyan, 2023; Vu et al., 2025). It appears that a shift is needed from a static view of trade to a dynamic understanding of structural transformation, moving from low-tech commodities to products with higher technological and innovation export intensity, allowing countries to capture more value within global value chains. In this context, the OECD industry classification (based on R&D intensity) provides a key framework for distinguishing between high-tech, medium–high-tech, and low-tech sectors. In this area, it is also essential to focus on the longitudinal integration of these flows at the national level. The scholarly literature lacks research that systematically tracks how changes in a country’s R&D structure and patent activities over time catalyze changes in its comparative advantages (RCA). This study addresses this gap by directly linking the innovation input–output cycle with sectoral export specialization, thereby providing a holistic view of how technological progress drives international competitiveness.

Despite extensive research into innovation systems and trade specialization, a significant theoretical and empirical discontinuity remains in the literature. While innovation performance models (input–output) primarily focus on the efficiency of transforming resources into knowledge, the trade specialization literature based on the RCA index often approaches export patterns from a static perspective. This study addresses this gap by integrating both streams. We argue that longitudinal tracking of the innovation cycle is essential for understanding the dynamics of structural transformation, where technological progress is not merely a byproduct but a direct determinant of shifts in nations’ comparative advantages.

Based on the theoretical framework and the identified research gap, the following hypotheses were formulated:

H1. 

There is a statistically significant positive relationship between R&D expenditure per capita and the number of patent applications per million inhabitants in the selected European countries, reflecting the efficiency of the National Innovation System.

H2. 

There is a statistically significant positive relationship between the number of scientists and engineers and innovation output (as measured by the number of patent applications), acting as the primary building blocks of intrapreneurial activity.

H3. 

Countries with higher innovation performance (measured by patents) exhibit a stronger shift in Revealed Comparative Advantage (RCA) toward medium- and high-technology intensity sectors, signaling successful structural transformation.

3. Materials and Methods

Patents, as a vital indicator of innovative capacity, serve as a key variable in numerous studies (Chu, 2021; Leogrande et al., 2022; Nie et al., 2023; Yang et al., 2024). Patent applications stimulate increased economic activity, which subsequently exerts a significant influence on economic growth and, consequently, national competitiveness (Yildiz & Görkey, 2024). In accordance with Jose et al. (2022), patenting is closely linked to the research and development sector, and its intensity depends on the number of researchers and their motivation to enforce intellectual property rights. Although patents are the most frequently used measure of innovation, Donoso (2017) argues that they may be a somewhat distorting indicator, as they do not always account for the quality of innovation, such as the export of technology-intensive products. According to Tadevosyan (2023), patent applications exhibit a causal relationship with exports. This assertion is further supported by Kittová and Družbacká (2023), who state that a development strategy focused on fostering innovation potential positively impacts the domestic value added of exported products, thereby enhancing the competitive position of any economy.

Following these findings, particularly those of Jose et al. (2022) and Tadevosyan (2023), the primary objective of this study is to identify the relationship between the number of EPO patent applications per million inhabitants and two variables: R&D expenditure per capita and the number of scientists and engineers per million inhabitants across four European countries (Czechia, Germany, Italy, and Slovakia). Concurrently, the study employs the Revealed Comparative Advantage index to identify their comparative advantages in the export of selected commodity groups originating from medium-to-high innovation-intensity industries. By achieving this objective, the objective is not only to identify the innovation performance of the surveyed countries and the link between patents and research sector indicators but also to determine whether the implementation of innovation contributes to their competitiveness and translates into the ability to export high-tech products.

The selection of the analyzed countries was intentional. Given that the Slovak Republic is primarily focused on automotive manufacturing—a sector classified as medium-innovation-intensive—it was of particularly interesting to evaluate its RCA position compared to other countries with a similar focus. Therefore, the comparative sample includes Slovakia, Czechia, Italy, and Germany. The primary data sources for this research were Eurostat and Trade Map (ITC) statistics. Selected indicators were analyzed for the period 2015–2024. This timeframe was chosen to provide a contemporary view of innovation dynamics within the EU and to ensure that the findings reflect the latest available data and global market trends.

Within the research, firstly, the normality of the sample sets was verified using the Shapiro–Wilk normality test, based on the following formula:

$$W={\displaystyle \frac{{({\sum }_{i=1}^m{a}_i^{\left(n\right)}(x_{n-i+1}^{*}-x_i^{*}))}^2}{{\sum }_{i=1}^n{(x_i-\overline{x})}^2}}$$

(1)

where x_i is the smallest number in the set; $\overline{x}$ is the mean value of the set; coefficients $a_i^{\left(n\right)}$ are table values; m is a vector, where $m={\displaystyle \frac{n}{2}}$, when n is even, respectively $m={\displaystyle \frac{n-1}{2}}$, when n is not even.

Upon confirming that all sample sets followed a normal distribution, we conducted a correlation analysis using the Pearson correlation coefficient:

$$r_{xy}={\displaystyle \frac{n{\sum }_{i=1}^n{x}_i{y}_i-{\sum }_{i=1}^n{x}_i{\sum }_{i=1}^n{y}_i}{\sqrt{\left[n{\sum }_{i=1}^n{x}_i^2-{\left({\sum }_{i=1}^n{x}_i\right)}^2\right]\left[n{\sum }_{i=1}^n{y}_i^2-{\left({\sum }_{i=1}^n{y}_i\right)}^2\right]}}}$$

(2)

where n is the number of observations; x is the independent variable; y is the dependent variable.

This allowed us to identify the magnitude and direction of the dependence between the examined indicators. The number of patent applications per million inhabitants was considered the dependent variable, while R&D expenditure per capita and the number of scientists and engineers per million inhabitants were the independent variables. To verify the statistical significance of the correlation results, we applied a significance test for the correlation coefficient:

$$t={\displaystyle \frac{r\sqrt{n-2}}{\sqrt{1-r^2}}}$$

(3)

where n is the number of observations, r is the correlation coefficient.

In addition, to verify the suitability of the chosen model, we also calculated the coefficient of determination r².

Finally, to assess the comparative advantages of the countries, we used the Revealed Comparative Advantage (RCA) index, based on the following formula:

$$RCA=({\displaystyle \frac{x_{ic}}{\sum x_{.c}}})/({\displaystyle \frac{x_{iw}}{\sum x_{.w}}}),$$

(4)

where

• $x_{ic}$ country c’s exports in commodity group i
• $\sum x_{.c}$ total exports of country c
• $x_{iw}$ world export w in commodity group i
• $\sum x_{.w}$ total world exports.

An RCA value greater than one indicates that a country specializes in the export of the commodity group. Conversely, a value less than one indicates the absence of a comparative advantage. For a more precise determination, the RCA values are categorized into four groups:

• 0 < RCA ≤ 1 absence of comparative advantage.
• 1 < RCA ≤ 2 small comparative advantage.
• 2 < RCA ≤ 4 medium comparative advantage.
• 4 < RCA great comparative advantage.

To ensure the validity of our findings and to account for the unobserved heterogeneity across the analyzed countries, we transitioned from individual correlation analyses to a Panel Data Regression model. Simple regression model and the model with Fixed Effects was employed to capture country-specific characteristics that remain constant over time. The panel regression model for assessing the relationship between research and development and other indicators and innovations was used in their research by, for example, Demir (2016) and Dritsaki and Dritsaki (2023).

In addition, recognizing that innovation is a time-dependent process, a lag (t − 2) was implemented, which accounts for the time difference between financial investments and the number of researchers and engineers on the one hand and the formal granting of patent rights on the other. Given the possibility of dependence between the selected dependent variables, a regression analysis of the impact of R&D expenditures on the number of patents granted was conducted (without examining the dependence on the number of researchers and engineers) being based on the following formula:

$$y_{it}=\alpha +{\beta }_1{x}_{1,i,t-1}+{\beta }_2{x}_{2,i,t-1}+{\epsilon }_{it}$$

(5)

where

• y_it—number of patent applications in country i in year t.
• x_1,i,t−2—R&D expenditure per capita (lag 2).
• x_2,i,t−2—number of scientists and engineers per million population (lag 2).
• α—constant.
• β₁, β₂—regression coefficients.
• ε_it—error term.

4. Results

The research part of this study, as stated in the methodology, first focuses on identifying innovation performance in four European countries (Czech Republic, Germany, Italy and Slovakia). The analysis then focuses on assessing whether and to what extent the identified innovation performance translates into the international competitiveness of produced goods, through the Revealed Comparative Advantage index in medium-to-high innovation-intensive sectors.

The results of the Shapiro–Wilk normality test (see Table 1) p (0.360; 0.651; 0.343) > α (0.05) indicate that for none of the variables examined (number of patent applications to the EPO per million inhabitants, R&D expenditure per capita, number of scientists and engineers per million inhabitants) there is no reason to reject the null hypothesis. This means that all three sample sets are normally distributed, and therefore, further calculations can be performed. The correlation coefficient between the number of patent applications to the EPO and R&D expenditure has taken the value of −0.382, which indicates a moderately strong negative tightness of the dependence between these variables. The negative dependence is because, despite the fact that in most of the monitored years there was an increase in investments in R&D, the number of patent applications to the EPO decreased year-on-year. Similarly, in the case of identifying the dependence between the number of scientists and engineers and the number of patent applications at the EPO, we can speak of the existence of a moderately strong negative tightness of dependence (−0.492). The significance test of the correlation coefficient 0.276 > α (0.05), or 0.149 > α (0.05) indicates that there is generally no statistically significant relationship between the number of patent applications at the EPO and investments in research and development, or the number of scientists and engineers, despite the demonstrated moderately strong negative tightness of dependence, and therefore, in Germany, the number of patent applications at the EPO depends on factors other than those examined here. This statement is also confirmed by the low value of the coefficient of determination (0.146, or 0.242), which shows that only 14.6%, or 24.2% of the variability in the number of patent applications at the EPO can be explained by changes in research investment or by changes in the number of scientists and engineers. Such results may indicate that the impact of innovation inputs on patent activity may manifest itself with a certain time lag (Gurmu & Perez-Sebastian, 2008), or that the number of scientists and engineers alone may not be a decisive variable (Bloom et al., 2020).

Table 1. Correlation analysis—Germany.

α = 0.05	Patent Applications to the EPO per Million Inhabitants	Research and Development Expenditure per Capita in €	Number of Scientists and Engineers per Million Inhabitants
2015	303.7	1093	37,421
2016	302.8	1122	39,041
2017	309.0	1206	39,531
2018	321.6	1264	39,463
2019	322.1	1325	40,493
2020	311.2	1282	41,315
2021	311.2	1354	44,868
2022	296.0	1459	45,146
2023	297.2	1588	46,459
2024	300.0	1624	48,332
Shapiro–Wilk normality test	0.360	0.651	0.343
Correlation coefficient	-	−0.382	−0.492
Correlation coefficient significance test	-	0.276	0.149
Coefficient of determination	-	0.146	0.242

Source: Eurostat (2026), own calculations.

Since the p-value of the Shapiro–Wilk normality test in Italy (see Table 2) α < p (0.561; 0.720; 0.649), we accept H0 and can conclude that the selected data are normally distributed, or that the difference between the sample and the normal distribution is not large enough to be statistically significant, so further calculations can be performed. When determining the relationship between the number of patent applications at the EPO and research and development expenditures, the correlation coefficient showed a value of 0.839, which indicates that there is a very strong positive linear relationship between the monitored variables, i.e., when research and development expenditures increase, the number of patent applications at the EPO also increases. The value of the significance test of the correlation coefficient, which is very low (0.002), shows that the relationship between the variables under investigation is not random, but statistically significant. This is also confirmed by the value of the coefficient of determination r² = 0.704, which shows that the model explained more than 70% of the total variability. Also, when examining the relationship between the number of scientists and engineers and the number of patent applications at the EPO, the correlation coefficient r = 0.832 showed a very strong positive relationship, the statistical significance of which was confirmed by the significance test of the correlation coefficient (0.003), or the correctness of the chosen model confirmed by the coefficient of determination r² = 0.692. Based on the findings, it can be assumed that if investments in research and development increase in Italy, and at the same time, the number of scientists and engineers in the economy increases, the number of patent applications filed at the EPO will also increase, and therefore, the country’s innovation performance itself. In Italy, unlike Germany, the variability in the number of patent applications at the EPO is sufficiently explained by changes in the variables under study.

Table 2. Correlation analysis—Italy.

α = 0.05	Patent Applications to the EPO per Million Inhabitants	Research and Development Expenditure per Capita in €	Number of Scientists and Engineers per Million Inhabitants
2015	65.6	365	17,153
2016	68.5	382	17,534
2017	72.0	393	17,640
2018	72.9	417	18,454
2019	74.8	439	19,349
2020	77.7	420	19,523
2021	83.2	448	19,167
2022	85.7	439	19,256
2023	86.1	498	20,747
2024	82.3	516	21,273
Shapiro–Wilk normality test	0.561	0.720	0.649
Correlation coefficient	-	0.839	0.832
Correlation coefficient significance test	-	0.002	0.003
Coefficient of determination	-	0.704	0.692

Source: Eurostat (2026), own calculations.

Before performing the correlation analysis, the normality of the distribution of individual variables was verified in the case of Slovakia (see Table 3), as in the previous countries, using the Shapiro–Wilk test. The p value for all three sample sets (0.452; 0.198; 0.754) > α (0.05), which means that there is no significant deviation from normality and, as in the previous cases, we accept H₀ about a normally distributed distribution and can continue with further calculations. The correlation coefficient expressing the dependence between research and development expenditures and the number of patent applications at the EPO acquired the value of 0.782. This is a large positive tightness of the dependence. The p-value of the significance test of the correlation coefficient 0.008 < α (0.05) says that the relationship between the variables under study is not random, but statistically significant, just like in Italy. The same is true in the second case, where the correlation coefficient expressing the dependence between the number of scientists and engineers and the number of patent applications at the EPO acquired a similar value of 0.780, which also expresses a large positive tightness of the dependence. The low value of the significance test of the coefficient 0.008 < α (0.05) shows that the detected relationships cannot be considered random and are valid at the chosen significance level. The coefficient of determination reached the value of 0.612 for research and development expenditures and 0.608 for the number of scientists and engineers, which means that approximately 61.2% of the variability in the number of patent applications can be explained by research and development expenditures and approximately 60.8% of the variability is explained by the number of scientists and engineers. The results in Slovakia, as well as the results in Italy, thus reflect the relatively great importance of investment and qualified human resources in shaping patent activity, and therefore, the innovation potential of these economies.

Table 3. Correlation analysis—Slovakia.

α = 0.05	Patent Applications to the EPO per Million Inhabitants	Research and Development Expenditure per Capita in €	Number of Scientists and Engineers per Million Inhabitants
2015	8.7	171	16,712
2016	8.1	118	16,420
2017	7.5	138	18,582
2018	9.4	138	20,282
2019	7.7	143	22,108
2020	9.9	154	24,478
2021	7.9	168	25,239
2022	10.3	198	27,269
2023	10.3	236	29,528
2024	11.4	236	34,914
Shapiro–Wilk normality test	0.452	0.198	0.754
Correlation coefficient	-	0.782	0.780
Correlation coefficient significance test	-	0.008	0.008
Coefficient of determination	-	0.612	0.608

Source: Eurostat (2026), own calculations.

Even in the case of the Czech Republic (see Table 4), the Shapiro–Wilk normality test did not show a significant deviation from normality p (0.316; 0.414; 0.375) > α (0.05), which means that all three sample sets have a normal distribution and can be the subject of further calculations. The results of the correlation analysis (r = 0.558) indicate a moderately strong positive relationship between the number of patent applications at the EPO and research and development expenditure per capita. However, the p-value of the significance test of the correlation coefficient 0.093 > α (0.05) shows that despite the existence of a moderately strong relationship between the variables mentioned, this relationship cannot be considered statistically significant. The coefficient of determination showed that only 31.1% of the variability in the number of patent applications is explained by changes in investment in research and development, and therefore, the remaining part of the variability must be explained by other factors not identified here. The correlation coefficient expressing the dependence between the number of patent applications at the EPO and the number of scientists and engineers reached the value r = 0.456, which is also a moderately strong positive tightness of the dependence. However, given the values of r, it could be expected that the significance test of the correlation coefficient 0.185 > α (0.05) would not demonstrate a statistically significant relationship in this case either, or the coefficient of determination (r² = 0.208) shows that the model explains only a small part of the total variability in the variable under study. In the Czech Republic, in both cases, we can speak of a positive tightness of the dependence between the variables under study, but its size is not sufficient to generally consider it relevant. Therefore, in the case of the Czech Republic, as in the case of Germany, it would be appropriate to identify another group of factors in the future that could explain a larger part of the variability in the number of patent applications at the EPO. The existence of these unidentified factors could be the key to a better understanding of innovation and research processes in the mentioned countries.

Table 4. Correlation analysis—Czech Republic.

α = 0.05	Patent Applications to the EPO per Million Inhabitants	Research and Development Expenditure per Capita in €	Number of Scientists and Engineers per Million Inhabitants
2015	20.2	308	30,650
2016	17.9	281	31,410
2017	19.4	325	31,563
2018	23.3	378	32,790
2019	19.0	408	34,048
2020	19.3	401	35,665
2021	19.1	453	38,381
2022	20.8	516	38,900
2023	22.0	538	39,224
2024	23.1	536	41,915
Shapiro–Wilk normality test	0.316	0.414	0.375
Correlation coefficient	-	0.558	0.456
Correlation coefficient significance test	-	0.093	0.185
Coefficient of determination	-	0.311	0.208

Source: Eurostat (2026), own calculations.

The correlation analysis performed showed that the relationships between the selected indicators of innovation performance (expenditure on research and development, number of scientists and engineers) and the number of patent applications to the EPO are not statistically significant enough. It can therefore be argued that the correlation analysis itself did not provide a sufficiently comprehensive picture of the mechanisms of innovation performance formation in the studied countries. The reason may be the relatively small size of the sample, but also the fact that the effects may manifest themselves with a time lag that the correlation analysis does not consider. For the reasons stated, in the next section, we will expand the empirical analysis performed by regression modeling using panel data. The lagged effect of research and development variables will be considered in the panel regression analysis, which will allow for a better identification of the causal relationships between innovation inputs and patent activity in the studied economies.

Panel regression analysis was used to capture the dynamic impact of R&D expenditure and the number of scientists and engineers on the development of the number of patent applications filed at the EPO in the countries under study (see Table 5). The model included lagged values (t − 2) of R&D expenditure per capita and the number of scientists and engineers per million population, thereby considering the fact that the effects of innovation inputs on patent activity may manifest themselves with a time lag. The analysis is based on 40 observations covering four countries over 10 years. The results show that there is a statistically significant positive impact of R&D expenditure on the number of patent applications filed at the EPO, as evidenced by a very low p-value of 0.000 ˂ α (0.05). The regression coefficient β reached a value of 0.336, which means that an increase in R&D expenditure per capita by one euro in the previous period is associated with an increase in the number of patent applications with a two-year lag of 0.336. It can therefore be stated that R&D investment plays an important role in shaping patent activity in the countries studied, with the effect of expenditure being manifested with a time lag. In the case of the number of scientists and engineers per million inhabitants, a weak negative, but statistically significant, effect on the number of patent applications at the EPO was identified. The regression coefficient β in this case took on a low negative value of −0.003, while the p-value of 0.000 ˂ α (0.05) was also extremely low in this case. This result, although statistically significant, suggests that the number of workers alone without adequate financial coverage or other qualitative factors may not directly generate a higher number of patents. This signals the need for future expansion of the model with additional explanatory variables (e.g., structure of research teams or cooperation with practice). The relatively high value of the coefficient of determination r² = 0.950 indicates that the model explained 95% of the total variability in the number of patent applications at the EPO, including their time lag. The value of the Durbin–Watson test (0.352) indicates the presence of the positive autocorrelation of the residuals, which is common in panel data, but the overall predictive power of the model remains high. Based on the results, it can be argued that panel regression analysis provides more convincing evidence of the existence of a relationship between innovation inputs and patent activity than simple correlation analysis. Research and development spending has a statistically significant positive impact on the development of the number of patent applications, while it is also important to consider the time lag. The importance of human capital, represented by the number of scientists and engineers, is present in the model, but its effect was negative and very weak. This result suggests that quantity alone, without considering qualitative factors, may not automatically lead to an immediate increase in patent production.

Table 5. Panel regression analysis.

Indicator	Number of Patents in the EPO per Million Inhabitants	Research and Development Expenditure per Capita in €	Number of Scientists and Engineers per Million Inhabitants
		Lag 2 (t − 2)	Lag 2 (t − 2)
Number of observations (n)	40 (4 countries/10 years)	40	40
Regression coefficient (β)	-	0.336	−0.003
Standard error	-	0.017	0.000
T-statistics	-	20.243	−4.303
p-value (significance)	-	0.000	0.000
Durbin–Watson test	0.352
Coefficient of determination (r2)	0.950

Source: Eurostat (2026), own calculations.

Based on the results obtained, the first two hypotheses set out in the theoretical part are to be evaluated:

• Hypothesis H1, that “in selected European countries there is a statistically significant positive relationship between R&D expenditure per capita and the number of patent applications per million inhabitants, which reflects the efficiency of the National Innovation System”, was accepted, as the regression coefficient β for R&D expenditure acquired a value of 0.336 when examined with a time lag at an extremely low p-value of 0.000.
• Hypothesis H2, that “there is a statistically significant positive relationship between the number of scientists and engineers and innovation output (measured by the number of patent applications), which act as the main building blocks of intra-business activity”, was rejected. Although this relationship is demonstrable based on the low p-value (0.000), the result contradicts the original assumption, as the regression coefficient β took on a very low, negative value of −0.003. This suggests that the increase in the number of scientists and engineers in the studied countries did not lead to an increase in patent activity but had a slightly negative impact on it.

Since in the previous part, the paper dealt with the patent applications as one of the key factors in the development of innovation, and the innovation ability of countries leads to the modernization of industry and thus increases the competitiveness of a country’s exports on the global market and therefore the overall competitiveness of the country (Hu et al., 2024), the impact of countries’ innovation ability on their competitiveness in the global economy was analyzed by identifying RCA for selected commodity groups of products produced in industrial sectors that largely depend on innovation performance. According to Zawislak et al. (2017), sectors with high-to-medium innovation intensity include, for example, aircraft and spacecraft production, the pharmaceutical industry, production of medical, optical and other precision instruments, the chemical industry, the automotive industry, etc. As part of the analysis, 5 commodity groups that are products of some of the above-mentioned sectors were identified, and through RCA, whether the countries studied specialize in the production of these products, or what a great comparative advantage they have was identified. Given that the Slovak Republic is known for its participation in the automotive industry, it is crucial to know how it would fare in the RCA compared to other countries that are also well-known car producers. And, for this reason, the compared sample of countries includes, in addition to Slovakia, the Czech Republic, Italy and Germany.

The RCA values reported in the four countries under study within the five selected commodity groups (see Table 6), which indicate whether a given country specializes in the production of the product group in question, show that there is a relatively significant sectoral specialization in the countries under study. In the period 2015–2024, Italy initially showed a small, and from 2022, already a medium comparative advantage within commodity group 30: Pharmaceutical products with an RCA value above 2.0, which indicates an increasing potential to compete on the global pharmaceutical market. In the monitored period, Germany showed values around 2.0 in this commodity group, i.e., also a small-to-medium comparative advantage. Although these values do not indicate significant dominance, given the innovative capabilities of the German economy, the potential for further development can be assumed. Neither the Czech Republic nor Slovakia, with RCA values of <1, show any comparative advantage in this area, which is among the most innovation-intensive, and therefore cannot compete with the largest producers.

Table 6. RCA for 2015–2024.

	2015	2016	2017	2018	2019	2020	2021	2022	2023	2024
30 Pharmaceutical products
Czech Republic	0.46	0.46	0.45	0.46	0.46	0.43	0.44	0.44	0.46	0.45
Italy	1.39	1.47	1.67	1.66	1.90	1.86	1.60	2.06	2.07	2.20
Germany	1.83	1.81	1.93	2.03	1.84	1.82	1.94	2.12	1.96	1.96
Slovakia	0.22	0.23	0.14	0.16	0.16	0.17	0.14	0.17	0.17	0.21
31 Fertilizers
Czech Republic	0.22	0.22	0.25	0.23	0.24	0.22	0.24	0.02	0.22	0.23
Italy	0.25	0.29	0.29	0.26	0.25	0.27	0.26	0.21	0.24	0.29
Germany	0.19	0.22	0.23	0.23	0.23	0.23	0.24	0.25	0.21	0.22
Slovakia	1.1	1.04	1.05	0.96	0.91	0.90	1.11	1.22	0.86	1.00
84 Nuclear reactors, boilers, machinery and mechanical appliances; parts thereof
Czech Republic	1.50	1.58	1.60	1.69	1.68	1.68	1.69	1.69	1.55	1.46
Italy	1.66	1.70	1.67	1.65	1.60	1.52	1.57	1.54	1.57	1.47
Germany	1.38	1.41	1.46	1.49	1.49	1.43	1.47	1.46	1.47	1.38
Slovakia	0.99	1.07	1.05	1.02	1.05	1.02	1.09	1.13	1.05	0.96
87 Vehicles other than railway or tramway rolling stock, and parts and accessories thereof
Czech Republic	2.34	2.50	2.55	2.54	2.52	2.58	2.63	2.74	2.56	2.68
Italy	0.99	1.01	1.05	1.04	0.97	1.05	1.09	1.08	0.99	0.89
Germany	2.18	2.17	2.16	2.14	2.05	2.09	2.21	2.42	2.15	2.18
Slovakia	3.18	3.39	3.28	3.78	4.01	4.72	4.68	4.73	4.39	4.42
90 Optical, photographic, cinematographic, measuring, checking, precision, medical or surgical instruments and apparatus; parts and accessories thereof
Czech Republic	0.49	0.55	0.61	0.65	0.65	0.61	0.68	0.81	0.82	0.88
Italy	0.73	0.76	0.74	0.76	0.76	0.72	0.83	0.92	0.88	0.89
Germany	1.37	1.45	1.54	1.61	1.64	1.60	1.67	1.79	1.71	1.75
Slovakia	0.29	0.26	0.34	0.33	0.31	0.28	0.30	0.34	0.33	0.32

Source: TradeMap (2026), own processing. Note: Colors indicate the size of the comparative advantage (green—large comparative advantage, red—no comparative advantage).

Within commodity group 31: Fertilizers, a different structure of comparative advantages is shown, with RCA values being relatively low in all countries. The Czech Republic, Italy and Germany showed RCA values < 1, which means that fertilizers do not represent their key export segment. Only Slovakia showed RCA > 1.0 in several years, which indicates that it had a relative comparative advantage in fertilizer production compared to the other monitored countries, but it was only small. It can therefore be stated that all the countries we studied have a relatively weaker position on the international market in commodity group 31: Fertilizers and do not reach a level that would indicate a significant specialization in the given commodity group.

In commodity group 84: Nuclear reactors, boilers, machinery and mechanical equipment, the Czech Republic, Italy and Germany achieved RCA values above 1.0 throughout the entire monitored period, which indicates a small comparative advantage. In Slovakia, RCA fell below 1.0 in 2015 and 2024, but not significantly. In most of the monitored years, Slovakia also showed a weak or marginal comparative advantage. It can therefore be said that all the analyzed countries have a relatively stable, but mostly low level of comparative advantage in commodity group 84, which indicates a similar degree of specialization in this area.

Since all countries included in the analysis are car producers, it could be expected that they would fare relatively well in the assessment of comparative advantages within commodity group 87: Vehicles other than railway vehicles, parts and accessories thereof. This was confirmed, with Slovakia showing a medium comparative advantage in vehicle production in the period 2015–2018. Since 2019, it has shown an RCA value of >4, which indicates a large comparative advantage, which is the result of the operation of several car manufacturers as well as several subcontractors of components for the automotive industry in the country. For the Czech Republic and Germany, we recorded medium comparative advantages in commodity group 87 with RCA values from 2.34 to 2.74 for the Czech Republic, or from 2.05 to 2.42 for Germany. Despite the fact that both countries are less specialized in car production than Slovakia, their potential to compete in the global car market is relatively large. On the contrary, Italy achieved RCA values of approximately 1, which means that it may have a relative comparative advantage in car exports compared to the rest of the world, but will continue to lag compared to the countries included in this analysis.

In the period 2015–2024, Germany also showed a small comparative advantage in commodity group 90: Optical, photographic, cinematographic, measuring, checking, precision, medical or surgical instruments and apparatus, which indicates a high technological level and strong innovation capacity in the field of precision instruments and medical technology. Germany is therefore the only country surveyed to have at least a small comparative advantage in 4 out of 5 possible areas. And although we did not record an RCA > 4, i.e., a large comparative advantage, for Germany in any of the commodity groups monitored, it had a larger number of small advantages. A country with moderate comparative advantages may signal that it is focusing on developing several sectors at the same time, as too much specialization in just one area can be risky. Italy, the Czech Republic and Slovakia showed RCA values < 1 in commodity group 90, which indicates that they are not focusing on exporting these products.

From the above, it follows that each of the monitored countries is specialized in certain areas and at the same time there are sectors in which it is less competitive. Slovakia is characterized by a large comparative advantage in the vehicle sector, which is associated with the extensive presence of the automotive industry in the country. It also shows one small comparative advantage, but that, like the large comparative advantage, comes from a sector with only medium innovation intensity (31: Fertilizers). In commodity groups 30 and 90, which represent sectors with high innovation intensity, Slovakia does not show any specialization. Germany shows a medium comparative advantage in one of the selected five commodity groups and a small one in the other three, with the potential to reach the level of medium comparative advantage in a short time, which indicates its flexibility and diversified approach to innovation-intensive sectors. In the case of Italy and the Czech Republic, we also identified certain comparative advantages related to the presence of the relevant industries in the territory of these countries, but their position slightly lags that of Germany. Furthermore, the Czech Republic, like Slovakia, shows comparative advantages only in sectors with medium innovation intensity. These differences reflect the different production capacities, levels of technological development and export strategies of the analyzed economies.

Based on the results obtained, we can proceed to the evaluation of the third hypothesis set out in the theoretical part:

• Hypothesis H3, that “countries with higher innovation performance (measured by patents) show a stronger shift in Revealed Comparative Advantage (RCA) towards sectors with medium and high technological intensity, which signals a successful structural transformation”, was confirmed. The analysis showed that Germany and Italy, which showed a higher number of patent applications to the EPO in the monitored period 2015–2024, can maintain and develop comparative advantages even in the most technologically demanding sectors. In contrast, Slovakia and the Czech Republic, with lower patent activity, show some specialization primarily in medium technologically demanding sectors.

5. Discussion

As part of the investigation of innovation performance, it was crucial to focus on patent applications as a priority, as they are often considered one of the key factors in the development of technological innovations, but also in the overall economic development of countries (Leogrande et al., 2022). They were analyzed in relation to two other indicators that contribute to the innovation performance of countries (expenditure on research and development, number of scientists and engineers). Correlation analysis in Italy in both cases revealed a strong positive correlation (r = 0.839, r = 0.832, respectively) between the examined indicators, which was statistically significant. A similar situation to that in Italy is also in Slovakia, where we also identified a relatively strong dependence (r = 0.782, r = 0.780) between the number of patent applications at the EPO and expenditure on research and development, respectively. In this case, too, the significance test of the correlation coefficient showed that these relationships are not random. It can be stated that the number of patent applications increases proportionally to the number of scientists and engineers, as well as the amount of research and development expenditure, which corresponds to the findings of Paula and Silva (2021), who claim that there is a positive impact of investment in research and development on national development, which is mediated by the total number of patent applications in the country. However, the situation in the remaining two monitored countries was different. In the Czech Republic (r = 0.558, r = 0.456), although a medium positive tightness of dependence was found among the analyzed indicators, it was statistically insignificant. In Germany, there was even a negative tightness of dependence (−0.382; −0.492), and these relationships were also not statistically significant. This phenomenon can be explained by taking into account the innovation maturity of countries: in Germany, as a highly developed innovation economy with a high number of scientists and consistently high research spending, the effect of additional investment on the number of patents may be less pronounced, while in less developed countries such as Slovakia or Italy, the marginal effects are higher. This is also pointed out by Kim (2018), according to whom the number of patents increases with better support for R&D, but the growth rate decreases at higher levels of investment.

The results indicate that in these two countries, other factors likely play a more significant role in shaping patent activity, which would be worth identifying in the future.

If we consider patents as one of the means of supporting innovation (Moser, 2013) and based on the previously mentioned finding by Paula and Silva (2021) or the study by Schmoch and Pouris (2021), which found that focusing on increasing the number of patent applications should contribute to improving the economic situation in the country, we could assume that the low number of patent applications and the statistically insignificant correlation between them and other determinants of innovation performance may be a brake on development in Germany or the Czech Republic. This is also indicated by some other studies, which add that in countries with lower GDP per capita, the number of patent applications is also lower (Diebolt & Hippe, 2019). However, this is not the case in Germany. Other research indicates that the patent system may not be an important part of development for low- and middle-income countries (Hall, 2020), but it may be other factors. From this perspective, the situation in Slovakia, where despite a strong positive correlation, patent activity is low, or in the Czech Republic, where the dependence is weaker, may not necessarily be perceived as explicitly negative. According to Levratto and Quignon (2020), patent costs are high and R&D expenditures may be directed to other areas, which may be one of the reasons why R&D investments in some countries do not correlate positively with the number of patent applications. However, whatever the reasons for this situation, it cannot be denied that in a country where R&D investments were higher (Germany), the total number of patent applications was also higher (Cherif et al., 2022). One would expect a similar conclusion to apply to the number of scientists and engineers in individual countries. However, in practice, we see that although the Czech Republic showed a much lower number of patent applications per million inhabitants compared to Germany throughout the entire monitored period, it lagged behind Germany only slightly in the number of scientists and engineers, and even surpassed Italy in this indicator, where the number of patent applications per million inhabitants is approximately 3 times higher than in the Czech Republic. Regarding the established dependence between the number of scientists and engineers and patent applications at the EPO in the period 2015–2024, a relatively strong and at the same time statistically significant relationship was demonstrated in Italy and Slovakia. This is in line with the findings of Huňady and Orviská (2014), who found that the number of patents in EU Member States is positively correlated with the number of employees in research and development, as growth in employment in this sector can be reflected in higher patent activity. However, Huňady and Orviská (2014) also point out that this mechanism may not work the same way in all countries, as is the case in Germany, where we found a negative correlation between the variables studied. This leads us to the conclusion that in Germany, as a highly developed country with a long-term high and stable number of scientists and engineers, their further increase may not lead to a proportional increase in the number of patent applications. To complement the correlation analysis, a panel regression analysis was conducted, which considered the time lag of innovation inputs. The results showed that R&D expenditure with a two-year time lag had a statistically significant positive impact on the number of patent applications in the countries studied in 2015–2024. This is in line with Hall et al. (2005), who found that R&D expenditure is a key determinant of patent activity, and that patents represent an important output of innovation processes. The number of scientists and engineers had a statistically significant weak negative impact on the number of patent applications, which was a surprise. It should be considered that not only is the quantitative aspect important, but the qualitative aspect of this indicator may play a greater role. For example, Bloom et al. (2020) also point out that the number of scientists is growing worldwide, but the number of innovations per scientist is decreasing. Based on the results, it can be concluded that panel regression provided more convincing evidence of the existence of a relationship between innovation inputs and patent activity in the studied economies than simple correlation analysis.

Since, according to Akyüz et al. (2020), sectoral strength at the micro level is a determinant of a country’s success at the macro level and at the same time it is shown that in order to achieve success at the international level, it is necessary to show certain comparative advantages; after conducting a correlation analysis, we focused on identifying the comparative advantages of the studied countries through RCA, focusing primarily on commodity groups produced in sectors with medium-to-high innovation intensity. According to De Melo and De Carvalho (2024), comparative advantages in sectors with medium- and high technological intensity indicate the ability of economies to add value to exports from domestic production, which means that economies must develop their innovation capabilities, while economies that do not show comparative advantages in sectors with high technological intensity tend to lag behind their competitors. In the context of this study, Germany can be therefore evaluated as the most successful, showing comparative advantages in most of the analyzed commodity groups, including sectors with high technological intensity 30: pharmaceutical production, 90: optical, photographic, cinematographic, measuring, checking, precision, medical or surgical instruments and apparatus. On the contrary, the Czech Republic and Slovakia achieve comparative advantages mainly in sectors with medium innovation intensity, which to some extent corresponds to their lower innovation performance measured by patent activity. Given the nature of the Czech and Slovak economies, this may also indicate that the country’s structural specialization may limit the country’s ability to develop comparative advantages in more technologically demanding sectors. Although this study does not allow us to identify a direct causal relationship between patent activity and comparative advantages in exports, the results indicate that countries with higher patent activity tend to show comparative advantages in more technologically intensive sectors. This finding is in line with the literature, which indicates that countries with a higher degree of innovation performance are more likely to experience productivity increases, which will indirectly lead to increasing their international competitiveness, especially in high value-added sectors (Tadevosyan, 2023). Based on the above, it can be concluded that supporting innovation performance is an important prerequisite for the long-term competitiveness of countries, especially in high value-added sectors.

The key scientific contribution lies in linking the examination of patent activity with the determinants of innovation performance (expenditure on research and development, number of scientists and engineers) and the subsequent connection to the comparative advantages of countries in sectors with medium-to-high innovation intensity. The originality of the approach is mainly based on the comparative analysis of four economies with different levels of development, which may point to differences in the functioning of the mechanisms of the innovation system depending on the level of economic development. The fact that the positive correlation between the number of patents in the EPO and the determinants of innovation performance is not manifested in the same way in all countries contributes to the consideration of the limit of the universality of patents as an indicator of innovation performance in the most developed economies (Germany). The study also seeks to identify the link between the level of patent activity and the ability of countries to achieve comparative advantages in technology-intensive industries, thereby expanding existing knowledge on the relationship between patent activity and international competitiveness by considering the existence of structural differences between Central European countries and more developed economies.

Theoretical and practical implications: Given the low number of patent applications in Slovakia and the Czech Republic, governments in these countries should focus on supporting patenting and increasing its efficiency, for example by simplifying the patent process or providing financial incentives. The efficiency of transforming R&D investments into patents could depend on the stage of economic development and the maturity of the country’s innovation system (developed vs. transforming economies), given the differences between countries. Therefore, it would be appropriate to consider structural specialization and the potential for patent saturation to use R&D inputs more efficiently. Given the situation in Germany, other determinants affecting patent activity should also be identified and supported, considering that the importance of patents as an indicator of innovation may differ between economically most advanced countries, such as Germany, and countries with a middle- or lower-income level, such as Slovakia or the Czech Republic. To increase the efficiency of innovation, Germany could consider moving from quantitative assessment of patents to value or qualitative indicators (e.g., patent citations, commercial success). Furthermore, since Slovakia and the Czech Republic mainly develop comparative advantages in areas requiring medium innovation, while Italy and Germany also in highly innovation-intensive sectors, lagging countries should also target and support sectors with high added value, which could contribute to increasing economic performance and competitiveness in them. Politically, this therefore points not only to supporting the diversification of the industrial structure, but, especially in Slovakia, also to a gradual reduction in the economy’s excessive dependence on the automotive industry.

6. Conclusions

The objective of this study was to identify the relationship between the number of patent applications to the EPO per million inhabitants and two other indicators in the field of innovation performance—the amount of research and development expenditure per capita, or the number of scientists and engineers per million inhabitants in four EU countries, and at the same time, through RCA, to identify their comparative advantages in the export of selected commodity groups of goods originating from sectors with medium-to-high innovation intensity. The set goal was achieved. The correlation analysis performed showed that the number of patent applications to the EPO correlates relatively strongly with two other indicators of innovation performance in Italy and Slovakia. However, the situation in these two countries is different. Although Italy significantly lags behind Germany in the number of patent applications, it also overtakes the Czech Republic and Slovakia. Slovakia, on the other hand, significantly lags behind all other countries studied. In Germany, the observed closeness of dependence was even negative, which suggests that the number of patent applications to the EPO in this country must depend on completely different factors than those examined here. A simple correlation analysis was supplemented by a panel regression. This provided more convincing evidence of the existence of relationships between innovation inputs and patent activity than correlation analysis. Research and development expenditure had a statistically significant positive effect on the development of the number of patent applications with a two-year time lag in the period under study, but the number of scientists and engineers with a two-year time lag showed a statistically significant, but slightly negative effect on patent activity. The impact of human capital on patent activity may therefore be determined by productivity rather than by their nominal increase. Looking at the indicators in individual countries in absolute numbers, Germany spends the most on research and development in absolute terms among the monitored countries, it also has the highest number of scientists and engineers per 100,000 inhabitants, and at the same time, it also shows the highest number of patent applications at the EPO. The second-highest number of patent applications was observed in Italy, which invests in research and development approximately as much as the Czech Republic, but in comparison, it shows half the number of scientists and engineers per 100,000 inhabitants. Slovakia shows the lowest values among the four countries compared in all indicators.

The results of the analysis examining the impact of innovation inputs on patent activity complement the view of the specialization of countries measured through RCA. Although the study does not examine a direct causal relationship between patents and RCA, from a comparative perspective certain parallels can be observed. Germany applies a diversified approach to innovation-intensive industries, developing specialization in the production of several commodities, including those from the most innovation-intensive industries. However, the comparative advantages arising from them are only at the level of small-to-medium-sized enterprises, which indicates that Germany’s potential in this area is present, but it has not yet reached the level of a significant competitive position on the global market or is facing intense pressure from global competition. Italy achieves a similar position to the Czech Republic, but with an important difference: its average comparative advantage comes from an area of high innovation intensity, unlike the Czech Republic. Slovakia and the Czech Republic lag in the most innovation-intensive industries and develop comparative advantages primarily in industries with only medium innovation intensity.

Although meeting the main objective set was managed, it is necessary to point out the fact that our research also has several limitations. In the first part, only two determinants were analyzed on which the development of the number of patent applications in individual countries may depend. However, to comprehensively assess the innovation performance of countries, it would be necessary to examine several other factors that may influence it. According to Taalbi (2025), patents capture only a part of the innovations implemented, because many innovations are not patented, so the use of patent indicators alone does not fully capture innovation performance. In the future, it would therefore be appropriate to supplement patent activity data with other indicators to obtain a more comprehensive picture of innovation processes in individual countries. In addition, a limitation is the fact that we used the standard version of RCA to identify comparative advantages, although some studies (Danna-Buitrago & Stellian, 2021; B. Liu & Gao, 2022) suggest that it is more appropriate to use extended RCA, which, in addition to exports, also takes into account GDP per capita, the trade balance of countries, etc. The aim of future research could therefore be to eliminate the inaccuracies resulting from the above facts.