Next Article in Journal
Research on the Role of Digital Finance in Urban Green Innovation
Previous Article in Journal
Spatial Pattern Evolution and Influencing Factors of Foreign Star-Rated Hotels in Chinese Cities
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
RSEE
Volume 2
Issue 1
10.3390/rsee2010002
rsee-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Social Filter Theory—A Sleeping Beauty of Regional Innovation Theories

by
Jianhui Ren
1,*,
Linlin Lai
2,
Binjie Pei
1 and
Wenyu Zhan
1
1
Institute of Transformation and Development of Resource-Based Economy, Shanxi University of Finance and Economics, Taiyuan 030006, China
2
Business School, Nanjing University, Nanjing 210008, China
*
Author to whom correspondence should be addressed.
Reg. Sci. Environ. Econ. 2025, 2(1), 2; https://doi.org/10.3390/rsee2010002
Submission received: 8 October 2024 / Revised: 12 December 2024 / Accepted: 18 December 2024 / Published: 31 December 2024
Browse Figures
Versions Notes

Abstract

As a branch of regional innovation theories, social filter theory fundamentally reveals the reasons behind the geography of innovation in different countries and regions. However, compared with other regional innovation theories, social filter theory has remained largely ignored and has not been fully developed. To enrich and develop social filter theory, this article systematically traces its origins and comments on four aspects: the concept of social filters, the construction and measurement of social filter indicators, the mechanisms, and empirical research on how social filters affect regional innovation and the transformation of innovation. Until recently, regional innovation theories primarily focused on superficially describing the phenomenon of innovation, but they paid little attention to the local social filter conditions necessary for innovation generation and transformation, which are prerequisites for constructing a mature innovation system. Therefore, further efforts are needed to address the current knowledge gap based on the recommended directions in this article.

1. Introduction

In the era of knowledge economy, innovation has become the main driving force for the development of every country and region. Innovation is becoming increasingly important not only for creating new market opportunities, but also for promoting economic transformation and upgrades, and for enhancing a country’s international competitiveness. Zoltaszek and Olejnik highlight that the literature on regional innovation research has advanced significantly over the past few decades [1], including regional innovation systems [2,3], learning regions [4,5], innovation milieus [6,7], regional innovation networks [8], regional innovation patterns [9,10], and social filter theory [11,12,13,14,15], among others. We searched for relevant keywords through the Web of Science to summarize the frequency of these theories (Figure 1). It is worth noting that studies on social filter theory have been relatively scarce, with research progress lagging behind the vigorous development of other innovation theories from 1999 to 2022. Consequently, social filter theory has not attracted widespread attention from scholars. However, social filter conditions are prerequisites for the successful development of all innovation activities, as they affect not only the generation of regional innovation, but also the efficiency of innovation transformation [16,17]. Thus far, there remains a lack of unified understanding regarding the conceptual scope and theoretical framework of the social filter, with terms such as “social filter” [14], “cultural filter” [18,19,20], and “institution playing the role of ‘social filter’” [21,22,23] frequently appearing in the literature.
In the above statement, social structures, cultures, and institutions are only certain dimensions or domains of the social filter. The synergy of these elements is crucial for building a mature innovation system. Scholars such as Rodríguez-Pose argue that different regions exhibit significant variation in their ability to transform R&D efforts into innovation and economic growth, due to the distinct social filters present in each region. Social filter theory emphasizes that regional socio-economic conditions constitute the social filter, which may have heterogeneous impacts on innovation and the ability to innovate and transform, forming invisible barriers to knowledge spillover between regions and affecting the spatial diffusion of knowledge [17,24,25]. Throughout the available literature on social filters, apart from most studies appearing as review citations, Rodríguez-Pose and his collaborators have conducted some empirical studies using social filters as an alternative tool for regional innovation systems, primarily focusing on developed regions, such as the EU, the United States, and Italy [21,26], as well as emerging economies like China, India, Russia, and Mexico [24,27,28,29].
In general, social filter theory provides strong explanations for the formation of innovation geography in various countries and regions, as well as the law of the direction and the extent of knowledge space diffusion. Over the past 20 years, it is unfortunate that social filter theory has remained dormant, like a sleeping beauty, and has not been fully developed. However, as innovation continues to gain importance in society, social filter theory is poised to gain renewed attention, ushering in a new era of academic exploration and making a significant contribution to the field of regional innovation theory.
This article is structured as follows. Section 2 summarizes the concept of social filters, tracing the relevant literature and identifying the connotation of social filters. In Section 3, we clarify the measurement indicators and social filter index. In Section 4, we illustrate the mechanisms through which the social filter influences regional innovation and its transformation, highlighting its role both within and across regions. In Section 5, we compare related empirical research based on social filters. Our discussion in Section 6, along with some perspectives, outlines broader future directions for the development of social filter theory.

2. What Is the Social Filter?

Since Schumpeter proposed the theory of innovation, it has become a widely accepted notion that technology serves as the engine of economic growth. However, it is not always a smooth process from R&D to innovation and then to economic growth. A social filter refers to the selective effect of a series of social, cultural, and institutional factors on the innovation process. It involves the mechanism and process of social acceptance, and the adoption and dissemination of new technology and new ideas. The local “social filter” may influence which innovations are adopted and take root in a particular region or cultural context. Social filters selectively promote or limit the diffusion of innovation. For example, cultural factors may influence people’s acceptance of new technologies, or local government policies may promote or limit innovation activities.
Based on this, Rodríguez-Pose (1999), building on the concept of relationship space [30], argues that each region has a unique social filter [31] composed of innovative and conservative elements that either promote or hinder the development of successful regional innovation systems [32,33,34]. If the innovative elements of a region’s social filter dominate, its ability to adopt technology and obtain returns from R&D investment will be stronger. Conversely, if the conservative elements dominate, the region’s ability to obtain returns from R&D will be weaker [35]. According to the social filter condition, regions can be divided into two types: innovation-prone regions and innovation-averse regions (see Figure 2). Regions with weak social filters, such as innovation-prone regions, exhibit adequate permeability to innovation and change. They are more open to external innovation and actively learn from advanced regions. These regions can smoothly transition from product innovation to economic activities and efficiently transform their innovation inputs into outputs [36]. On the other hand, the social filters of innovation-averse regions are almost impervious. Companies and even entire societies in these countries and regions lack the ability to transform technological innovation into higher value-added economic activities. It can be observed that the social filter creates both resistance and permeability effects between R&D investment and economic activity. The social filter consists of regional socio-economic and institutional conditions, along with intangible capital such as human and social capital, which collectively influence knowledge generation, diffusion, and, ultimately, innovation performance [37,38,39].
The coexistence of innovation-prone regions and innovation-averse regions suggests that not all regions possess the same capacity for innovation and transforming it into economic growth [40,41]. In some situations, some regions experience significant economic growth as a result of their R&D investment, while others derive less benefit from their investment despite the same R&D investment. This is because social filters create impedance and permeation effects between R&D investment and economic activities in each region. Social filters are not only a “negative barrier“, they may also act as a “positive catalyst“, enhancing innovation capacity by increasing regional absorption of external knowledge and innovation. Thus, social filters have a dual function: they are both a driver of innovation and a potential constraint on it. Similar to RIS, social filter theory recognizes R&D activities as the only cognitive dimension [1].
On the one hand, a considerable portion of the skilled and unskilled blue-collar workers, who dominated the era of mass production, were trained according to Taylor’s principles, making it difficult to redeploy them and adapt to new technological transformations. On the other hand, a rigid labor market has struggled to provide suitable jobs for the younger generation of talent. The lower social mobility and limited economic activities in certain social sectors have further contributed to the transformation of these regions into innovation-averse societies. Similarly, some peripheral regions struggle with insufficient innovation dynamics, making it difficult to achieve growth. Additionally, these peripheral regions face challenges related to weak economic organization and a lack of entrepreneurial motivation.
In this regard, the concept of a social filter provides a more convincing explanation for the diverse regional innovation patterns observed in different countries and regions. A social filter is essentially a system composed of local social conditions, and due to the incoordination of various factors, R&D activities cannot fully penetrate into the production system. The maturity of a region’s innovation system is often influenced by its social filter. A social filter that supports and limits innovation may directly affect whether a region is able to establish a sound innovation system. Of course, it is not one single socio-economic factor in isolation that matters for innovation. Instead, it is the combination of a set of local features—human capital, young people, social capital, and other factors—that captures their synergies and interactions. In extreme cases, it is challenging to expect innovation to emerge in societies characterized by fragile trust, human capital deficiencies, and inadequate protection of property rights. A virtuous cycle of human capital accumulation can only emerge when human capital combines with a youthful population structure and a vibrant labor market [42].

3. Indicators and Measurement Methods of Social Filters

A social filter is the result of a multifactor composite, and its complexity and intertwined characteristics make it challenging to describe qualitatively and quantitatively, hampering comparative analysis between regions [43]. In order to measure social filters, scholars have continuously explored how to specifically characterize their indicators, leading to the establishment of rigorous conceptual frameworks and practical measurement approaches [16,24].

3.1. Key Indicators for Measuring Social Filters

Currently, there is still debate regarding the measurement indicators of social filters among domestic and international scholars. In the early stages of research on social filters, social filters could be measured by indicators including demographic structure, agricultural labor proportion, long-term unemployment rate, and education level, serving as the prototype indicators of social filters [14,32]. Moreover, socio-economic indicators should reflect a region’s skill levels, labor market conditions, and economic structure [44]. Among them, human capital is a core indicator of a social filter [43,45]. This is because human capital, as reflected by education levels, can represent a region’s capacity for learning and absorption, and many observable and unobservable factors (e.g., talent mobility, local education quality, and the social returns of education) can indirectly impact innovation through the accumulation of human capital. In particular, the variables that seem to be more relevant for shaping the social filter are those relative to three aspects, namely educational attainment, productive employment of human resources, and demographic structure [15]. Specifically, compared to other sectors, productive employment in the agricultural sector tends to have lower productivity, which often represents hidden unemployment. The long-term unemployment rate is an important indicator of labor market rigidity, reflecting the plight of regional workers who lack the necessary skills for productive work. Young people are the driving force behind regional innovation and social change, and regions with active youth populations tend to generate more innovation [46]. Ultimately, the above aspects were synthesized in a preliminary framework of social filter indicators, including local market rigidity, demographic structure, education, technology, and human capital, as well as scientific infrastructure [47]. These socio-economic conditions can act as both catalysts, by promoting innovation, and filters, by impeding it [48,49].
In addition to economic structural elements, scholars have also incorporated cultural and institutional factors into the system of social filter indicators (see Table 1). For example, entrepreneurial culture should be factored into the social filter [50]. Social exclusion and institutional efficiency should also be included besides the core indicator of human capital when the primary concern is the institutional level of society [26]. For one thing, regional innovation depends on the sharing and exchange of information or knowledge among individuals. Adverse social institutional conditions can hinder the local economic system’s ability to absorb knowledge and innovation. For another, regional innovation is also constrained by social capital and institutions, impeding economic growth [28]. Empirical research demonstrated that institutional quality plays a role in the social filter, affecting a region’s ability to transform innovation into economic growth [51]. Additionally, institutions play a crucial role in shaping the innovation environment of firms, either facilitating or hindering learning and knowledge spillovers, thereby influencing their innovation performance [21]. Other scholars further refine the institutional indicators, including urbanization rate, social capital, privatization, financial development index, and property rights development index. Furthermore, the agglomeration effect of urbanization promotes innovation, while social capital supports it by facilitating the dissemination of knowledge and valuable information. Institutional factors, such as contract law, property rights, and privatization, also play critical roles in fostering innovation [52].

3.2. Social Filter Index

A social filter, as proposed by the aforementioned scholars, is composed of a unique set of social and structural elements. Some of these elements promote regional economic activity, while others hinder it [21]. Moreover, there is correlation and even information overlap among certain constituent indicators. Including all of these variables simultaneously in a regression model leads to collinearity. Therefore, principal component analysis (PCA) is used to measure social filters by synthesizing these variables into a single indicator that can capture as much of the variation in the original indicators as possible. By combining these variables into a single indicator, principal component analysis provides a “joint measure” for assessing the social filter in each region.
Scholars often use the first principal component as the social filter index [55,56,57,58] to measure the social filter conditions across different regions. In the research conducted on EU member states, the first principal component explained 43.1% of the total variance, with an eigenvalue significantly greater than 1. The signs of the weight coefficients for the sub-variables align with expectations: the impact of the educated population, educated labor force, and participation in lifelong learning is positive, while the influence of agricultural labor and long-term unemployment is negative, and the weight for the 15–24 age group is relatively small [15]. In research on the states in the United States, the first principal component alone is able to account for around 36 percent of the total variance, and the proportion of the population aged 15–24 and the proportion of the population with a bachelor’s degree and higher had negative contributions to the comprehensive social filter index, while the proportion of the unemployment rate and agricultural employment population had positive contributions [42]. This is contrary to the findings of EU member states [15]. In comparative research on China and India [27], the first principal component alone accounts for 45% and 36% of the total variance in the original variables considered for China and India, respectively. The weight coefficients of the sub-variables were completely opposite in China and India. The influence of the proportion of young people and the proportion of the population that is educated is negative in China and positive in India; the impact of unemployment and the proportion of people employed in primary industries is negative in India and positive in China. It is not difficult to find that China’s result is the same as the measurement result of Rodríguez-Pose [47] for American states, while the result for India is consistent with the results of EU member states.
In the research on Mexican states, the first principal component used to measure the social filter explained 54% of the total variance. Each sub-variable made the expected positive or negative contribution to the composite variable of the social filter index. Educational and skill variables, as well as the percentage of the young population, were positively correlated with the composite social filter index, while the percentage of agricultural employment showed a negative correlation [24]. Similarly, in Chinese prefecture-level cities, the first principal component explained 37.8% of the total variance. However, the weights for the proportion of the young population and private sector employees were negative [53]. This contrasts with the positive contribution of the youth population to the social filter index observed in the EU and Mexico. Results from other scholars are not listed here.
It can be observed that although using principal component analysis (PCA) to measure the social filter and taking the first principal component score as a social filter index have been widely accepted and utilized by scholars, the results of this index measurement can vary. These variations stem from differences in research subjects and the selection of constituent indicators. Notably, there is a lack of rigorous scientific justification and explanation for choosing the indicators that compose the social filter. The issues of generalization and arbitrariness in indicator selection urgently need to be addressed. Thus, establishing scientific evaluation criteria and striving for a unified understanding are critical. Moreover, in statistics, aside from PCA, other methods such as cluster analysis, analytic hierarchy process (AHP), and self-organizing mapping (SOM) are also commonly employed to condense high-dimensional data into low-dimensional representations. Exploring these alternative methods may provide valuable insights for improving the measurement of the social filter index.

4. Mechanisms of Social Filters on Regional Innovation and Transformation

For regional innovation and innovation transformation, the mechanism of the social filter can be understood as a process of digestion and absorption, where the constituent elements of the social filter form a filter organization [59]. In terms of filter mechanisms, the social filter manifests itself in two ways, internal filter and external filter, corresponding to intra-regional R&D and inter-regional knowledge spillovers, respectively. In terms of filter channels, the social filter influences the generation, dissemination, absorption, and utilization of knowledge, ultimately impacting the effectiveness of regional innovation and innovation transformation. In order to comprehensively summarize and reveal the mechanisms through which the social filter affects regional innovation [60] and innovation transformation, this research presents a conceptual diagram illustrating the interplay between the social filter and knowledge spillovers in two regions (Figure 3). For instance, consider an economy with two regions, A and B, where Region A has more adequate social filter conditions than Region B. The darker region is a knowledge production department, referring to universities and research institutions, while the lighter region is an innovation transformation sector, referring to the production department. In this way, the baffle inside the ball in Figure 3 is similar to the social filter generated by the region itself on knowledge transformation in Figure 2, and the surface of the ball depicts the social filter generated by the two regions on knowledge spillover from both sides. As can be seen from Figure 3, Region B has a stronger social filter effect on both regional knowledge transformation and inter-regional knowledge spillover. Its production department can absorb knowledge from its own and its neighbor’s investment due to an adequate social filter condition in Region A. Due to deficient social filter conditions in Region B, its production department cannot fully utilize inter-regional and intra-regional knowledge spillovers.

4.1. Heterogeneous R&D Activities and Innovation Transformation

The R&D process can be divided into two stages: knowledge creation and innovation transformation [61]. Knowledge possesses the characteristics of a public good, with both non-rivalrous and partially excludable attributes [62]. Therefore, the location of innovation transformation can be completely different from the place of innovation [54]. Knowledge generated within a region and knowledge generated outside the region are both important sources of regional growth [63]. In fact, regions can choose to simultaneously transform local innovations and innovations from other regions into local production activities, thereby promoting local economic growth [64]. However, other factors should also be taken into account, as different regions demonstrate varying capacities in absorbing and transforming innovation into economic growth.
R&D activities in different regions are not homogeneous. They exhibit various types due to differences in product characteristics and ownership forms, resulting in disparities in productivity [65]. Private-sector R&D investments are typically more directly linked to the development of new products and services. In contrast, research conducted by the public sector and higher education institutions often focuses on basic research, which presents challenges for the commercialization of inventions and patents [44]. This disconnect, where new knowledge and creative ideas struggle to be effectively translated into product or process innovation, is referred to as the “knowledge paradox” [66].
In addition, different types of R&D activities lead to varying economic performances, and the selection of regional innovation patterns should be aligned with the region’s development stage and conditions. For example, if a region is at a relatively low stage of development, even a strong emphasis on independent innovation and increased R&D investment may fail to yield significant results due to a lack of the necessary talent pool and market foundation to support the advancement of cutting-edge technologies [67]. The choice of innovation patterns should align with a country or region’s factor endowments, institutional environment, and technological ecosystem. Only the right match can effectively achieve sustained high-level technological progress. Bilbao-Osorio and Rodríguez-Pose [44] also highlight the numerous obstacles faced by peripheral regions in terms of R&D investments. Firstly, financially constrained peripheral regions struggle to allocate sufficient funding to reach the critical mass for innovation due to the high costs of R&D activities. Secondly, these regions often lack a well-established innovation system involving government, industry, university, and research institutions, and they have limited capacity to establish technological linkages with other regions. Lastly, peripheral regions often lack a clear strategy for scientific and technological development. In contrast to more developed regions, Oughton further emphasize that peripheral regions have a higher demand for innovation inputs. Nonetheless, peripheral regions have relatively low capacity to absorb public funds for promoting and investing in innovation-related activities.
However, it is worth noting that the challenges faced by peripheral regions in achieving innovation through independent R&D investments may extend beyond these bottlenecks. It may seem more reasonable for underdeveloped regions to achieve development by joining other innovation core regions. In reality, there is no single standard for how different regions should choose their preferred innovation patterns, and further exploration is needed.

4.2. Social Filter and Regional Internal R&D Activities

R&D activities are the main form of regional innovation, and social filter conditions vary across regions. The social filter, composed of specific social, political, and economic features, influences the ability to transform R&D investments into innovation and economic growth. According to Rodríguez-Pose’s [14] description of the social filter, a stronger social filter indicates worse local social filter conditions. The alignment between regional R&D activities and social filter conditions may result in different economic performances. Figure 4 provides a visual representation of the potential combinations between regional R&D investments and social filter conditions, categorizing regions into four types: innovation-averse regions, innovation “fetter regions”, innovation-prone regions, and innovation-leading regions [68].
As shown in Figure 4, the innovation-leading regions are located in the upper right corner. These regions have a strong permeability of social filter on R&D activities, and high levels of R&D investment can effectively drive economic growth. Innovation-leading regions are typically the innovation centers and growth poles of a country or region, acting as exemplars of economic and social development. In contrast, innovation-averse regions are located in the lower left corner of the figure. These regions have insufficient R&D investment and a worse social filter. They often have low organizational thickness, lack motivation for innovation, and exhibit inertia and disregard for innovation. These regions are typically economically lagging and geographically peripheral. In addition to these two opposite situations, two intermediate scenarios may also exist. The innovation-prone regions, located in the lower right corner of the figure, have adequate social filter conditions but relatively low levels of R&D investment. These regions can achieve innovative development more easily by encouraging indigenous innovation efforts or absorbing knowledge spillovers from other regions, accumulating capital for R&D investment, and surpassing the threshold for R&D investment. Conversely, the innovation fetter regions in the upper left corner are less optimistic. Due to deficient social filter conditions, blindly encouraging R&D investment poses risks similar to “building a cathedral in the desert” [69]. Hence, the innovation transformation path indicated by the curved black arrows in the figure represents a more feasible and pragmatic approach.
It is evident that the economic potential of innovation activities in a region can only be realized when the region has adequate social filter conditions [70]. Specifically, the impact of these conditions on regional innovation can be explained from three main domains: educational achievements, productive employment of human resources, and demographic structure. Professionals and skilled labor are important forces for ensuring innovative mechanisms, and R&D becomes profitable when human capital reaches a certain threshold level. The employment rate is an important factor influencing the innovation process because young people possess stronger learning and adaptability skills, as well as a more innovative spirit and risk-taking attitude. The economic structure of a region significantly influences its ability to generate and absorb innovation, with agricultural regions being less likely to produce patents due to the sector’s lower innovativeness compared to high-tech industries. Conversely, certain sub-industries within manufacturing and service sectors may be more conducive to promoting innovation [44]. Moreover, the improvement of geographical accessibility and the accumulation of human capital can interact with regional innovation activities, facilitating the effective transformation of regional innovation into economic growth. Improvement in accessibility can influence the creation and dissemination of technology through various mechanisms. Accessibility serves as a good indicator reflecting the impact of distance on innovation because it considers not only the localized spillover represented by the frequency of interactions within a region but also the opportunities and barriers for communication between regions. Human capital accumulation, as the primary mechanism for innovation diffusion, highlights the interaction between education and innovation in the approach of the innovation system. The innovation system approach emphasizes the impact of education on the learning ability of local society [43].
These analyses have important policy implications for innovation in peripheral regions. A more effective approach to an innovation-based development strategy is not only to actively encourage R&D investment but also to focus on optimizing social filter conditions [16]. Furthermore, social filter conditions will shape the opportunity space for the future development of the region [71]. For many regions, improving social filter conditions is more challenging than increasing R&D investment, as these conditions do not improve automatically. This is because some components of social filter adjustments cannot be achieved in the short term. In the absence of automatic adjustment mechanisms in the labor market, regional disparities in technological labor endowments may persist [72]. At the national level, cultural and institutional barriers are common challenges faced by regions [46,73], and this is one of the significant reasons why only a few regional innovation centers exist in many countries. Therefore, comprehensive improvement of social filter conditions in various regions is an important measure to enhance innovation capacity and achieve a balanced innovation development method for every country.

4.3. Social Filter and Inter-Regional Knowledge Spillovers

The social filter plays a crucial role in the process of knowledge spillover. Knowledge spillover refers to the dissemination and transfer of knowledge from one region to another. However, the effectiveness of this knowledge transfer is often influenced by the conditions of the social filter. Factors such as the level of education, the strength of scientific research institutions, and the degree of cultural openness in a region significantly impact its capacity to absorb and utilize knowledge. In regions with weak social filter conditions, the ability to effectively absorb and apply knowledge is hindered, even when innovations from other regions “spillover”.
Although knowledge exchange can occur between regions, each region utilizes the capabilities of other regions differently. Even more concerning, local knowledge may not be effectively absorbed and utilized locally, but instead serve as a source of innovation for other regions. There is not always a direct connection between regional innovation and its growth. Sometimes, innovation in one region leads to increased income and employment in another region [74]. This type of knowledge spillover phenomenon is referred to as a reverse spillover [75]. Reverse spillovers indicate that a region’s innovation capacity and transformation capacity are not necessarily proportional. The frog-jumping phenomenon of “blooming within walls but fragrant outside”—where innovations benefit others more than their origin regions—is indeed present in reality. The Sicilian Paradox described serves as an example. Sicily, a lagging region in southern Italy, is not lacking in innovative companies, especially with STMicroelectronics ranking first in terms of patent authorizations among top Italian firms. However, these patents are not effectively utilized locally [76]. When the utilization capacity outside the region is strong, but local utilization capacity is weak, a region’s R&D and innovation efforts end up benefiting the development of other regions. The prerequisite socio-economic conditions for innovation must be established locally rather than relying on neighboring regions [42]. While every region seeks to utilize external innovation outputs, varying social filter conditions lead to differences in their capacities for innovation utilization. The social filter is the structural precondition of successful innovation development for regions. Therefore, social filter conditions not only affect knowledge production but also influence knowledge utilization [24]. From the perspective of exchange patterns, input–output mechanisms, and intellectual exchange, regions with weak social filter conditions are more likely to benefit from external knowledge spillovers [76]. Thus, while emphasizing knowledge spillovers, scholars should pay greater attention to the externalities of social filters [16].
In general, regarding regional economic growth, social filter conditions play a dual role as filters for innovative activities and invisible barriers to knowledge spillovers [77]. Regions attempting to achieve innovative development by free-riding face risks. If a region has poor social filter conditions, it not only fails to absorb knowledge from other regions but also struggles to diffuse its own knowledge to benefit others. The excessive concentration of knowledge or innovation in a few innovation centers leads to a deepening disparities between regions [78]. Therefore, for regions to gain an advantage in achieving economic growth through innovation, they should strive to improve their social filter conditions. Nonetheless, because of path dependence, adjusting or improving social filter conditions is not easily accomplished in the short term. In addressing this issue, some scholars suggest adopting a reverse thinking approach to consider measures or channels that can penetrate, overcome, and even remove the invisible barriers of social filters [79]. When the peripheral region is usually composed of a few large enterprises, there is a lack of local buzz, and at the same time, the absorption of knowledge spillovers from the core region is limited by geographical distance decay and deficient social filter conditions. In this regard, high-tech enterprises in periphery regions will deploy their R&D departments to core regions, and even, in some regions, the government can purchase land in core areas, establish industrial parks, and form innovation enclaves. In order to build knowledge pipelines, avoid knowledge filtering, and absorb knowledge from the core region, these measures aim to ensure that knowledge flows back to enterprises in peripheral regions. These measures try to absorb the knowledge dividend of core regions and gradually cultivate a local innovation system to overcome the negative impact of the local unfavorable social filter (Figure 5). Therefore, regions need to have the absorptive capacity to benefit from these knowledge spillovers or from a so-called “social filter” [25,80,81].

5. Empirical Analysis of Social Filter’s Impact on Regional Innovation and Knowledge Spillovers

Scholars have used modeling and empirical analysis to uncover how the social filter affects a region’s ability to absorb innovation and transform it into economic growth. Research in this field continues to be enriched and evolve. Initially, the analysis focused primarily on European countries, before expanding to include developed countries like the United States and Japan, and later extended to developing countries such as China, India, and Mexico. The analytical approaches employed include graphical methods, two-step analysis, and three-step progression methods. The econometric techniques have evolved from simple graphical representations to encompass simple linear regression, multiple linear and nonlinear regressions, instrumental variable methods, and dynamic GMM estimation. The analytical perspective has also shifted from initially providing a simple description of socio-economic conditions to analyzing externalities and their effects on economic growth, as well as the mechanisms of moderation and spillover effects.

5.1. The Relationship Between R&D and Economic Growth as the Starting Point

Schumpeter’s research highlights the threshold effect of R&D, along with the assumption of diminishing marginal returns in the neoclassical growth model. Building on this, Rodríguez-Pose [14] conducted a cross-sectional regression analysis of R&D investment and per capita GDP in Western European countries. The research found a significant positive correlation between R&D investment in technologically peripheral regions and economic growth, while the relationship between R&D investment and economic growth in technologically advanced regions was less apparent. However, the reliability of these results is questionable due to omitted variable bias, which makes it difficult to establish a direct causal relationship between R&D investment and economic growth, as other factors may be influencing the outcome. Rodríguez-Pose [14] also found that adverse socio-economic conditions, including low female employment rates, an aging workforce, insufficient high-skilled talent, and labor market rigidity, hinder innovation in peripheral regions. Simple regressions of R&D on patents or economic growth are undermined by endogeneity issues, casting doubt on the reliability of related research conclusions.

5.2. The Analysis Models and Methods

Represented by Rodríguez-Pose and his collaborators, the following scholars have expanded and enriched the models and methods that examine the impact of social filters on regional innovation and innovation transformation. Among these, this model is widely regarded as the most classic and frequently used approach [15,47]. The specific equation is as follows:
1 T l n ( Y i , t Y i , t − T ) = α + β 1 l n ( Y i , t − T ) + β 2 R & D i , t − T + β 3 S o c f i l t e r i , t − T + β 4 W R & D i , t − T       + β 5 W S o c F i l t e r i , t − T + β 6 W P G D P i , t − T + ε i , t − T
According to the growth model [82], when β 1 is negative, it indicates conditional convergence, meaning that the economic growth of different regions tends to converge. W is the spatial weight matrix; β 4 , β 5 , and β 6 represent knowledge spillovers, social filter spillovers, and economic spillovers, respectively. Among them, when the social filter has a positive spillover effect, it indicates that the behavior of free-riding may be acceptable; when the social filter spillover effect is negative, it indicates that there is a siphon effect and polarization may easily occur. This model integrates R&D inputs, local social filter conditions, and various external spillover effects in the innovation process. These factors collectively shape the formation and evolution of innovation systems, resulting in diverse regional innovation patterns across countries and regions. The model has been widely adopted in empirical studies, with enhancements incorporating moderation effects and threshold effects. In terms of estimation methods, scholars have commonly employed panel fixed effects, generalized method of moments (GMM), maximum likelihood estimation (MLE), Hausman–Taylor estimator, and so on [24,53,54,83].

5.3. The Drivers of the Geography of Innovation

The spatial agglomeration of innovation activities reflects the maturity of a country’s innovation system, and countries with a more balanced spatial distribution of innovation activities tend to have more mature national innovation systems [27]. So far, research on the geography of innovation, particularly in relation to the social filter, has mainly focused on TRIAD economies (the United States, European Union, and Japan) and BRIC economies (Brazil, Russia, India, and China). The United States is a global leader in innovation, with its most innovative regions located along the Eastern and Western coasts, as well as the Great Lakes region [42]. In the European Union, innovation activities are highly concentrated, with the 20 EU-15 regions accounting for about 70% of its total patents. For developing and emerging countries, Brazil’s innovation hubs are primarily centered around São Paulo, which brings together manufacturing activities, modern services, financial institutions, headquarters of large national and multinational subsidiaries, excellent infrastructure, and urban facilities [84]. Similarly, innovation in Russia is geographically concentrated, with eleven regions, including Moscow, St. Petersburg, and the Republic of Tatarstan, recognized as strong innovators [54]. In India, high-tech hubs such as Bangalore, Chennai, Delhi, Hyderabad, Mumbai, and Pune dominate the geography of innovation, accounting for 60–70% of the country’s patents [27]. In China, the spatial concentration of innovative activity is even more pronounced, with Guangdong, Beijing, and Shanghai producing 60% of the country’s patents [85]. Mexico, the 11th largest economy in the world and the largest emerging country outside the BRIC economies, also exhibits significant regional differences in innovation capacity. Eight states—Mexico State, the Federal District, Veracruz, Jalisco, Puebla, Guanajuato, Chiapas, and Nuevo León—account for 50% of the nation’s total R&D expenditure [24].
In addition to the high concentration of innovation activities, there are significant differences in innovation drivers in different countries and regions (Table 2). In the European Union, the United States, India, and Mexico, R&D investment is the primary determinant of innovation. Adequate social filter conditions enhance the returns on R&D, indicating better human capital endowment, lower unemployment rates, and a younger population structure that promotes knowledge generation and dissemination. In the European Union, India, and Mexico, knowledge spillovers also play a positive role in fostering innovation and the conversion of outcomes. The most innovative regions in these countries are those that provide adequate socio-economic conditions and are able to absorb, digest, and utilize R&D and knowledge spillovers. However, compared to developed regions in the European Union and the United States, as well as emerging countries like India and Mexico, China’s innovation dynamics rely more on agglomeration forces rather than traditional innovation drivers such as R&D investment, human capital endowment, and knowledge spillovers. The spatial range of knowledge spillover radiation from China’s major innovation sources is limited to metropolitan regions and has not achieved inter-regional diffusion [85]. To sum up, the country’s path to innovation is unique and diverse. As the old saying goes, all roads lead to Rome.

6. Future Perspectives Toward a Research Agenda

Equally important to theories on regional innovation systems, learning regions, innovative milieu, and regional innovation networks, social filter theory has emerged as a significant branch within regional innovation theory. These theories attempt to identify the prerequisites for regional innovation and explain the link between regional innovation and economic performance. The difference lies in the emphasis of the former four theories on the importance of proximity and inter-relationships among innovation actors. In contrast, social filter theory highlights how the socio-economic conditions of different regions generate impedance and permeation effects on R&D investment and knowledge spillovers, ultimately affecting innovation transformation. When the impedance effect outweighs the permeation effect, a strong filter occurs, thereby weakening the economic returns on R&D investment and knowledge spillovers. Conversely, when the impedance effect is smaller than the permeation effect, a weak filter occurs, enhancing the economic returns on R&D investment and knowledge spillovers. In summary, social filter theory is an important step in empirically researching the role of the regional innovation system. It also provides a new way to explain the heterogeneity of regional innovation and transformation capacity, as well as the asymmetry of inter-regional knowledge spillovers. Objectively speaking, there are still some shortcomings in the theoretical framework and practical application of social filter theory that need to be addressed and improved. Therefore, we recommend the following directions for future research.
Firstly, the concept and the index system of a social filter need to be further clarified. According to the idea conceived by the proponents of the concept of a social filter, each nation and region has a unique social filter that hampers or sustains innovation [63]. Its permeability reflects the region’s ability to absorb innovation and transform it into economic activity. This is actually a relatively abstract concept, so when scholars portray social filters, there are different descriptions of social conditions, socio-economic conditions, regional innovation systems, social institutions [26], social filter conditions [86], etc. At the same time, scholars have not sufficiently defined the connotation of a social filter, and its constituent elements have not yet reached a consensus. Sometimes they are even simply regarded as a collection of economic structural variables [87]. Although terms such as “cultural filter”, “social filter”, and “institution plays the role of a ‘social filter’” appear in different studies, we believe that social structure, culture, institutions, and other aspects are merely dimensions or domains of the social filter. In order to avoid over-generalizing the concept, the above description can be unified under the concept of a social filter, which is more conducive to the construction and development of social filter theory [68].
Secondly, future research should put more emphasis on the mechanisms through which social filters influence regional innovation and knowledge spillover. Currently, the existing research on mechanisms primarily focuses on the interaction between a social filter and internal regional innovation, as well as internal regional knowledge spillovers. However, this research is still at a general and descriptive stage. There is a lack of detailed research and understanding of how social filters affect the absorption, digestion, and utilization of knowledge, how the strength or weakness of permeability is manifested, and why it may act as an invisible barrier to knowledge spillovers. In particular, there is a lack of in-depth analysis of the characteristics and mechanisms of the constituent elements of social filters. Moreover, we should pay attention to the externality of social filters and explore the mechanisms and channels of their spatial spillover.
Thirdly, there is a need to improve empirical methods and models that examine the impact of the social filter on regional innovation and knowledge spillovers. Currently, existing empirical studies have conducted general tests [88]. However, the process through which the social filter influences the transformation of regional innovation into economic growth is complex. The social filter can have both facilitating and inhibiting effects, and a simplistic testing approach cannot effectively distinguish between them. Additionally, promoting regional innovation for economic growth involves a series of transformations, from R&D investment to innovation, production, and, ultimately, economic growth. However, most existing studies focus on independent and localized analyses, lacking a comprehensive examination of the entire process. Furthermore, these studies mostly rely on multiple regression models, with insufficient consideration of endogeneity.
Fourthly, the literature has so far been dominated by case studies of developed countries or regions, such as the EU and the United States. Whether this is applicable to East Asian and African countries requires further exploration. Another step would be to clarify the content filtered by the social filter, whether it is filtering redundant knowledge or digesting and absorbing beneficial knowledge. Regarding the spatial scale of social filters, on the one hand, different regions have different social filters, so what is the minimum spatial scale of the region? On the other hand, it remains unclear whether regions should be divided by their social filters, or whether a social filter should be used to divide regions. In either case, similar regions would share the same social filter. Moreover, it remains to be explored whether the concept of the social filter can transcend the regional scale, exist within enterprise organizations, and be influenced by their organizational structures and cultures.
In conclusion, the study of social filter theory can help us understand that innovation is not a simple technical process, but a complex social process. Different social and cultural environments affect the generation, diffusion and transformation of innovation through various mechanisms. An in-depth study of the theory and practice of social filters cannot only provide a new perspective for regional innovation research, but also provide valuable guidance for policy makers.

Author Contributions

Conceptualization, J.R.; formal analysis, J.R.; resources, L.L., B.P. and W.Z.; writing—original draft preparation, J.R.; writing—review and editing, J.R. and L.L.; visualization, L.L.; supervision, J.R.; project administration, J.R.; funding acquisition, J.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by The National Natural Science Foundation of China (Project number: 42101179) and the Humanities and Social Science Project of the Ministry of Education (Project number: 21YJC790096).

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Żółtaszek, A.; Olejnik, A. Regional effectiveness of innovation: Leaders and followers of the EU NUTS 0 and NUTS 2 regions. Innov. Eur. J. Soc. Sci. Res. 2024, 37, 399–420. [Google Scholar] [CrossRef]
  2. Cooke, P.; Uranga, M.G.; Etxebarria, G. Regional innovation systems: Institutional and organisational dimensions. Res. Policy 1997, 26, 475–491. [Google Scholar] [CrossRef]
  3. Iammarino, S. An evolutionary integrated view of regional systems of innovation: Concepts, measures and historical perspectives. Eur. Plan. Stud. 2005, 13, 497–519. [Google Scholar] [CrossRef]
  4. Healy, A.; Morgan, K. Spaces of innovation: Learning, proximity and the ecological turn. Reg. Stud. 2012, 46, 1041–1053. [Google Scholar] [CrossRef]
  5. Morgan, K. The learning region: Institutions, innovation and regional renewal. Reg. Stud. 2007, 41, 147–159. [Google Scholar] [CrossRef]
  6. Breschi, S.; Lissoni, F. Localised knowledge spillovers vs. innovative milieux: Knowledge “tacitness” reconsidered. Pap. Reg. Sci. 2001, 80, 255–273. [Google Scholar] [CrossRef]
  7. Camagni, R. The concept of innovative milieu and its relevance for public policies in European lagging regions. Pap. Reg. Sci. 1995, 74, 317–340. [Google Scholar] [CrossRef]
  8. Piazza, M.; Mazzola, E.; Abbate, L.; Perrone, G. Network position and innovation capability in the regional innovation network. Eur. Plan. Stud. 2019, 27, 1857–1878. [Google Scholar] [CrossRef]
  9. Camagni, R.; Capello, R. Regional innovation patterns and the EU regional policy reform: Towards smart innovation policies. Growth Chang. 2013, 44, 355–389. [Google Scholar] [CrossRef]
  10. Capello, R.; Lenzi, C. Regional innovation patterns from an evolutionary perspective. Reg. Stud. 2018, 52, 159–171. [Google Scholar] [CrossRef]
  11. Casi, L.; Resmini, L. Foreign direct investment and growth: Can different regional identities shape the returns to foreign capital investments? Environ. Plan. C Politics Space 2017, 35, 1483–1508. [Google Scholar] [CrossRef]
  12. Zhang, M.; Rodríguez-Pose, A. Government reform and innovation performance in China. Pap. Reg. Sci. 2024, 103, 100023. [Google Scholar] [CrossRef]
  13. Capello, R.; Lenzi, C. The dynamics of regional learning paradigms and trajectories. J. Evol. Econ. 2018, 28, 727–748. [Google Scholar] [CrossRef]
  14. Rodríguez-Pose, A. Innovation prone and innovation averse societies: Economic performance in Europe. Growth Chang. 1999, 30, 75–105. [Google Scholar] [CrossRef]
  15. Rodríguez-Pose, A.; Crescenzi, R. Research and development, spillovers, innovation systems, and the genesis of regional growth in Europe. Reg. Stud. 2008, 42, 51–67. [Google Scholar] [CrossRef]
  16. Crescenzi, R.; Rodríguez-Pose, A. Innovation and Regional Growth in the European Union; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2011. [Google Scholar]
  17. Rodríguez-Pose, A.; Crescenzi, R. Mountains in a flat world: Why proximity still matters for the location of economic activity. Camb. J. Reg. Econ. Soc. 2008, 1, 371–388. [Google Scholar] [CrossRef]
  18. Putnam, R.D. Making Democracy Work. Civil Traditions in Modern Italy; Princeton University Press: Princeton, NJ, USA, 1993. [Google Scholar]
  19. Tubadji, A.; Nijkamp, P. Six degrees of cultural diversity and R&D output efficiency: Cultural percolation of new ideas: An illustrative analysis of Europe. Lett. Spat. Resour. Sci. 2016, 9, 247–264. [Google Scholar]
  20. Borgoni, R.; Michelangeli, A.; Pontarollo, N. How Does a City Benefit from Culture? Evidence from Milan. Arts Adm. Mus. Stud. eJ. 2016, 1, 1–31. [Google Scholar] [CrossRef][Green Version]
  21. D’Ingiullo, D.; Evangelista, V. Institutional quality and innovation performance: Evidence from Italy. Reg. Stud. 2020, 54, 1724–1736. [Google Scholar] [CrossRef]
  22. Fratesi, U. Regional Knowledge Flows and Innovation Policy: A Dynamic Representation. Reg. Stud. 2015, 49, 1859–1872. [Google Scholar] [CrossRef]
  23. Akçomak, İ.S.; Müller-Zick, H. Trust and inventive activity in Europe: Causal, spatial and nonlinear forces. Ann. Reg. Sci. 2018, 60, 529–568. [Google Scholar] [CrossRef]
  24. Rodríguez-Pose, A.; Peralta, E.M.V. Innovation and Regional Growth in Mexico: 2000–2010. Growth Chang. 2015, 46, 172–195. [Google Scholar] [CrossRef]
  25. Faggian, A.; Partridge, M.; Malecki, E. Creating an environment for economic growth: Human capital, creativity or entrepreneurship? Int. J. Urban Reg. Res. 2017, 41, 997–1009. [Google Scholar] [CrossRef]
  26. D’Agostino, G.; Scarlato, M. Innovation, socio-institutional conditions and economic growth in the Italian regions. Reg. Stud. 2015, 49, 1514–1534. [Google Scholar] [CrossRef]
  27. Crescenzi, R.; Rodríguez-Pose, A.; Storper, M. The territorial dynamics of innovation in China and India. J. Econ. Geogr. 2012, 12, 1055–1085. [Google Scholar] [CrossRef]
  28. Smith, N.; Thomas, E. Socio-institutional environment and innovation in Russia. J. East-West Bus. 2015, 21, 182–204. [Google Scholar] [CrossRef]
  29. Kaneva, M.; Untura, G. Innovation indicators and regional growth in Russia. Econ. Chang. Restruct. 2017, 50, 133–159. [Google Scholar] [CrossRef]
  30. Massey, D. Spatial Divisions of Labour: Social Structure and the Geography of Production; Red Globe Press: London, UK, 1984. [Google Scholar]
  31. Fonseca, M. Southern Europe at a Glance: Regional Disparities and Human Capital. In Regional Upgrading in Southern Europe; Springer International Publishing: Cham, Switzerland, 2017; pp. 19–54. [Google Scholar]
  32. Sleuwaegen, L.; Boiardi, P. Creativity and regional innovation: Evidence from EU regions. Res. Policy 2014, 43, 1508–1522. [Google Scholar] [CrossRef]
  33. Ferretti, M.; Parmentola, A. Local Innovation Systems in Emerging Countries. In Local Innovation Systems in Emerging Countries; Springer: Cham, Switzerland, 2015; pp. 7–36. [Google Scholar]
  34. Crescenzi, R.; Rodríguez-Pose, A. Systems of Innovation and Regional Growth in the EU: Endogenous vs. External Innovative Activities and Socio-Economic Conditions. In Growth and Innovation of Competitive Regions; Springer: Berlin/Heidelberg, Germany, 2009; pp. 167–191. [Google Scholar]
  35. Ege, A.; Ege, A.Y. How to create a friendly environment for innovation? A case for Europe. Soc. Indic. Res. 2019, 144, 451–473. [Google Scholar] [CrossRef]
  36. Bramanti, A.; Fratesi, U. The Dynamics of an ‘Innovation Driven’ Territorial System. In Growth and Innovation of Competitive Regions; Springer: Berlin/Heidelberg, Germany, 2009; pp. 59–91. [Google Scholar]
  37. Lee, N. Psychology and the Geography of Innovation. Econ. Geogr. 2017, 93, 106–130. [Google Scholar] [CrossRef]
  38. Peiro-Palomino, J. The geography of social capital and innovation in the European Union. Pap. Reg. Sci. 2019, 98, 53–73. [Google Scholar] [CrossRef]
  39. Chen, A.; Li, Y.; Ye, K.; Nie, T.; Liu, R. Does transport infrastructure inequality matter for economic growth? Evidence from China. Land 2021, 10, 874. [Google Scholar] [CrossRef]
  40. D’Agostino, L.M.; Moreno, R. Green regions and local firms’ innovation. Pap. Reg. Sci. 2019, 98, 1585–1609. [Google Scholar] [CrossRef]
  41. Crescenzi, R. Undermining the Principle of Concentration? European Union Regional Policy and the Socio-economic Disadvantage of European Regions. Reg. Stud. 2009, 43, 111–133. [Google Scholar] [CrossRef]
  42. Crescenzi, R.; Rodríguez-Pose, A. R&D, Socio-economic conditions, and regional innovation in the U.S. Growth Chang. 2013, 44, 287–320. [Google Scholar]
  43. Crescenzi, R. Innovation and regional growth in the enlarged Europe: The role of local innovative capabilities, peripherality, and education. Growth Chang. 2005, 36, 471–507. [Google Scholar] [CrossRef]
  44. Bilbao-Osorio, B.; Rodríguez-Pose, A. From R&D to innovation and economic growth in the EU. Growth Chang. 2004, 35, 434–455. [Google Scholar]
  45. Fallah, B.; Partridge, M.D.; Rickman, D.S. Geography and High-Tech Employment Growth in US Counties. J. Econ. Geogr. 2013, 14, 683–720. [Google Scholar] [CrossRef]
  46. Crescenzi, R.; Rodríguez-Pose, A.; Storper, M. The territorial dynamics of innovation: A Europe-United States comparative analysis. J. Econ. Geogr. 2007, 7, 673–709. [Google Scholar] [CrossRef]
  47. Rodríguez-Pose, A.; Comptour, F. Do clusters generate greater innovation and growth? An analysis of European regions. Prof. Geographer. 2012, 64, 211–231. [Google Scholar] [CrossRef]
  48. Mosconi, F.; D’Ingiullo, D. Institutional quality and innovation: Evidence from Emilia-Romagna. Econ. Innov. New Technol. 2023, 32, 165–197. [Google Scholar] [CrossRef]
  49. Fratesi, U.; Senn, L. Regional Growth, Connections and Economic Modelling: An Introduction. In Growth and Innovation of Competitive Regions; Springer: Berlin/Heidelberg, Germany, 2009; pp. 3–27. [Google Scholar]
  50. Beugelsdijk, S. Entrepreneurial culture, regional innovativeness and economic growth. J. Evol. Econ. 2007, 17, 187–210. [Google Scholar] [CrossRef]
  51. Rodriguez-Pose, A.; Di Cataldo, M. Quality of government and innovative performance in the regions of Europe. J. Econ. Geogr. 2015, 15, 673–706. [Google Scholar] [CrossRef]
  52. Xiong, A.; Xia, S.; Ye, Z.P.; Cao, D.; Jing, Y.; Li, H. Can innovation really bring economic growth? The role of social filter in China. Struct. Chang. Econ. Dyn. 2020, 53, 50–61. [Google Scholar] [CrossRef]
  53. Rodríguez-Pose, A.; Zhang, M. Government institutions and the dynamics of urban growth in China. J. Reg. Sci. 2019, 59, 633–668. [Google Scholar] [CrossRef]
  54. Kaneva, M.; Untura, G. The impact of R&D and knowledge spillovers on the economic growth of Russian regions. Growth Chang. 2019, 50, 301–334. [Google Scholar]
  55. Rodríguez-Pose, A.; Ketterer, T.D. Do local amenities affect the appeal of regions in Europe for migrants? J. Reg. Sci. 2012, 52, 535–561. [Google Scholar] [CrossRef]
  56. Rodríguez-Pose, A.; Comptour, F. Evaluating the Role of Clusters for Innovation and Growth in Europe. In Geography, Institutions and Regional Economic Performance; Springer: Berlin/Heidelberg, Germany, 2013; pp. 209–235. [Google Scholar]
  57. Crescenzi, R.; Giua, M. The EU Cohesion Policy in context: Does a bottom-up approach work in all regions? Environ. Plan. A Econ. Space 2016, 8, 2340–2357. [Google Scholar] [CrossRef]
  58. Crescenzi, R.; Pietrobelli, C.; Rabellotti, R. Regional strategic assets and the location strategies of emerging countries’ multinationals in Europe. Eur. Plan. Stud. 2016, 24, 645–667. [Google Scholar] [CrossRef]
  59. Lee, N. Are innovative regions more unequal? Evidence from Europe. Environ. Plan. C Gov. Policy 2011, 29, 2–23. [Google Scholar] [CrossRef]
  60. Jan, F.; Feldman, M.P.; Martin, S. Technological dynamics and social capability: US states and European nations. J. Econ. Geogr. 2014, 14, 313–337. [Google Scholar]
  61. Crescenzi, R.; Rodríguez-Pose, A. An ‘integrated’ framework for the comparative analysis of the territorial innovation dynamics of developed and emerging countries. J. Econ. Surv. 2012, 26, 517–533. [Google Scholar] [CrossRef]
  62. Fragkandreas, T. When innovation does not pay off: Introducing the “European regional paradox”. Eur. Plan. Stud. 2013, 21, 2078–2086. [Google Scholar] [CrossRef]
  63. Capello, R.; Caragliu, A.; Nijkamp, P. Territorial capital and regional growth: Increasing returns in knowledge use. Tijdschr. Voor Econ. En Soc. Geogr. 2011, 102, 385–405. [Google Scholar] [CrossRef]
  64. Fujita, M. Towards the new economic geography in the brain power society. Reg. Sci. Urban Econ. 2007, 37, 482–490. [Google Scholar] [CrossRef]
  65. Griliches, Z. Productivity, R&D and basic research at firm levelin the 1970s. Am. Econ. Rev. 1986, 76, 141–154. [Google Scholar]
  66. Knockaert, M.; Spithoven, A.; Clarysse, B. The knowledge paradox explored: What is impeding the creation of ICT spin-offs? Technol. Anal. Strateg. Manag. 2010, 22, 479–493. [Google Scholar] [CrossRef]
  67. Sterlacchini, A. R&D, higher education and regional growth: Uneven linkages among European regions. Res. Policy 2008, 37, 1096–1107. [Google Scholar]
  68. Ren, J.; Lai, L. Social filter theory: A review of regional innovation theory. Reg. Econ. Rev. 2023, 63, 145–155. [Google Scholar]
  69. Qin, C.; Ren, J. Research progress on the relationship between social filter and economic growth. Econ. Dyn. 2016, 667, 115–123. [Google Scholar]
  70. Crescenzi, R.; Rodríguez-Pose, A. Infrastructure and regional growth in the European Union. Pap. Reg. Sci. 2012, 91, 487–513. [Google Scholar] [CrossRef]
  71. Kurikka, H.; Kolehmainen, J.; Sotarauta, M.; Nielsen, H.; Nilsson, M. Regional opportunity spaces—Observations from Nordic regions. Reg. Stud. 2022, 56, 1440–1452. [Google Scholar] [CrossRef]
  72. Riggi, M.R.; Maggioni, M.A. Regional growth and the co-evolution of clusters: The role of labour flows. In Growth and Innovation of Competitive Regions: The Role of Internal and External Connections; Springer: Berlin/Heidelberg, Germany, 2009; pp. 245–267. [Google Scholar]
  73. Augier, M.; Guo, J.; Rowen, H. The Needham Puzzle Reconsidered: Organizations, Organizing, and Innovation in China. Manag. Organ. Rev. 2016, 12, 5–24. [Google Scholar] [CrossRef]
  74. Shearmur, R.; Bonnet, N. Does local technological innovation lead to local development? A policy perspective. Reg. Sci. Policy Pract. 2011, 3, 249–270. [Google Scholar] [CrossRef]
  75. Wan, Q.; Yuan, L.; Yao, Z.; Zeng, L. The impact of R & D elements flow and government intervention on China’s hi-tech industry innovation ability. Technol. Anal. Strateg. Manag. 2023, 35, 857–874. [Google Scholar]
  76. Caragliu, A.; Nijkamp, P. The impact of regional absorptive capacity on spatial knowledge spillovers: The Cohen and Levinthal model revisited. Appl. Econ. 2012, 44, 1363–1374. [Google Scholar] [CrossRef]
  77. Capello, R.; Caragliu, A.; Fratesi, U. Breaking Down the Border: Physical, Institutional and Cultural Obstacles. Econ. Geogr. 2018, 94, 485–513. [Google Scholar] [CrossRef]
  78. Gertler, M.S.; Wolfe, D.A.; Garkut, D. No place like home? The embeddedness of innovation in a regional economy. Rev. Int. Political Econ. 2000, 7, 688–718. [Google Scholar] [CrossRef]
  79. Sotarauta, M. Policy learning and the ‘cluster-flavoured innovation policy’ in Finland. Environ. Plan. C Gov. Policy 2012, 30, 780–795. [Google Scholar] [CrossRef]
  80. Brenner, T.; Broekel, T. Methodological Issues in Measuring Innovation Performance of Spatial Units. Ind. Innov. 2011, 18, 7–37. [Google Scholar] [CrossRef]
  81. Crescenzi, R.; Rodríguez-Pose, A. Reconciling top-down and bottom-up development policies. Environ. Plan. A Econ. Space 2011, 43, 773–780. [Google Scholar] [CrossRef]
  82. Barro, R.J.; Sala-I-Martin, X. Economic Growth; MIT Press: New York, NY, USA, 1995. [Google Scholar]
  83. Ren, J. Innovation, Social Filter, and Regional Economic Growth; Shanxi People’s Publishing House: Taiyuan, China, 2021. [Google Scholar]
  84. Goncalves, E.; Almeida, E. Innovation and spatial knowledge spillovers: Evidence from Brazilian patent data. Reg. Stud. 2009, 43, 513–528. [Google Scholar] [CrossRef]
  85. Rodríguez-Pose, A.; Wilkie, C. Putting China in perspective: A comparative exploration of the ascent of the Chinese knowledge economy. Camb. J. Reg. Econ. Soc. 2016, 9, 479–497. [Google Scholar] [CrossRef]
  86. Rabellotti, R.; Crescenzi, R.; Pietrobelli, C. Innovation drivers, value chains and the geography of multinational corporations in Europe. J. Econ. Geogr. 2014, 14, 1053–1086. [Google Scholar]
  87. Paci, R.; Marrocu, E. Knowledge Assets and Regional Performance. Growth Chang. 2013, 44, 228–257. [Google Scholar] [CrossRef]
  88. Jesse, S.; Yawo, M.; Godwin, A. Determinants of regional economic resilience and post-crisis growth path performance: Insights from Canada. Reg. Stud. 2024, 12, 1–15. [Google Scholar]
Rsee 02 00002 g001
Figure 1. Number of relevant studies from 1999 to 2022. Source: Web of Science. Note: Among them, the literature on social filter theory is manually organized.
Figure 1. Number of relevant studies from 1999 to 2022. Source: Web of Science. Note: Among them, the literature on social filter theory is manually organized.
Rsee 02 00002 g001
Rsee 02 00002 g002
Figure 2. Relationship between R&D investment and economic activities in innovation-prone regions and innovation-averse regions. Source: Based on the literature [14].
Figure 2. Relationship between R&D investment and economic activities in innovation-prone regions and innovation-averse regions. Source: Based on the literature [14].
Rsee 02 00002 g002
Rsee 02 00002 g003
Figure 3. A schematic diagram illustrating the internal and external filters within regions. Source: Author’s elaboration.
Figure 3. A schematic diagram illustrating the internal and external filters within regions. Source: Author’s elaboration.
Rsee 02 00002 g003
Rsee 02 00002 g004
Figure 4. R&D investment and social filter. Note: The arrows in the diagram represent the transition from innovation-averse regions to innovation-leading regions, characterized by increased R&D investment and improved social filter conditions. Source: Modified based on the literature [34].
Figure 4. R&D investment and social filter. Note: The arrows in the diagram represent the transition from innovation-averse regions to innovation-leading regions, characterized by increased R&D investment and improved social filter conditions. Source: Modified based on the literature [34].
Rsee 02 00002 g004
Rsee 02 00002 g005
Figure 5. Knowledge spillover, pipeline, and social filter. Source: Author’s elaboration.
Figure 5. Knowledge spillover, pipeline, and social filter. Source: Author’s elaboration.
Rsee 02 00002 g005
Table 1. Social filter indicator systems in relevant studies.
Table 1. Social filter indicator systems in relevant studies.
AuthorSocial Filter ProxySecondary IndicatorTertiary Indicator
Rodríguez-Pose (1999) [14]Social conditionsDemographic structure—
Agricultural labor
Unemployment rate
Education level
Bilbao-Osorio and Rodríguez-Pose (2004) [44]Socio-economic factorsLevel of skillPercentage of adult population (25–59 years old)
Labor market situationEmployment rate
Economic structurePercentage of employees in high-tech manufacturing and service industries
Crescenzi(2005) [16]Human capitalEducational attainmentPercentage of educated population
Crescenzi et al. (2007) [46]; Rodríguez-Pose and Crescenzi (2008) [15]; Crescenzi et al. (2012) [27]; Rodríguez-Pose and Peralta (2015) [24]Socio-economic conditions Educational achievementsLifelong learning
Labor force with higher education
Productive employment of human resourcesPercentage of labor force employed in agriculture
Long-term unemployment rate
Demographic structurePercentage of youth population (15–24 years old)
Rodríguez-Pose and Comptour D’Agostino (2012) [47]Socio-economic conditionsLocal market rigiditiesLong-term unemployment rate
Agriculture employment
Corporate tax rate
Demographic aspectsPercentage of youth population (15–24 years old)
Education, skill, and human capitalTotal population education
Lifelong learning
Scientific base of the regionHuman resources in science and technology
D’Agostino and Scarlato (2015) [26]Social institutional conditionsSocial exclusionLong-term unemployment rate
Juvenile unemployment rate
Family poverty index
Education levelHigh school dropout rate at the end of the first year
Secondary education rate
Percentage of employed adults
Institutional efficiencyMunicipal waste-sorting services
Perception of the risk of crime
Rodríguez-Pose and Zhang (2019) [53]Social filterDemographic structurePercentage of youth population (15–24 years old)
Sectoral compositionPercentage of agricultural employment
Use of human resourcesEmployment rate
Ownership structureShare of employment in private firms
Kaneva and Untura (2019) [54]Social–economic filterAvailability of a skilled labor forceShare of university graduates
Share of labor with tertiary education
Demographic structureEmployed labor force aged 15–30
Industrial structureShare of labor force employed in agriculture
Xiong et al. (2020) [52]Social filter conditionsUrbanization rateProportion of residents living in cities
Social capitalNumber of social organizations per ten thousand people
PrivatizationPercentage of private fixed investment
Financial development indexChina’s Marketization Index Report (2011)
Property rights development indexChina’s Marketization Index Report (2011)
Table 2. Comparison of innovation systems across countries and regions.
Table 2. Comparison of innovation systems across countries and regions.
AuthorPeriodCountries and RegionsThe Drivers of the Geography of InnovationDegree of Innovation Spatial Agglomeration
R&D investmentKnowledge spilloverSocial filterSocial filter spilloverAgglomeration effect
Crescenzi et al. (2012) [27]1995–2007ChinaNoneNoneNoneNegativePositiveHigh
1995–2004IndiaPositivePositivePositiveNonePositiveHigh
Kaneva and Untura (2019) [29]2005–2013RussianPositiveNoneNegativeNone—High
Rodríguez-Pose and Peralta (2015) [24]2000–2010MexicoPositivePositivePositivePositive—High
Crescenzi et al. (2007) [46]1990–2002EUNonePositivePositiveNonePositiveLow
1990–1999USAPositiveNonePositiveNonePositiveLow
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ren, J.; Lai, L.; Pei, B.; Zhan, W. Social Filter Theory—A Sleeping Beauty of Regional Innovation Theories. Reg. Sci. Environ. Econ. 2025, 2, 2. https://doi.org/10.3390/rsee2010002

AMA Style

Ren J, Lai L, Pei B, Zhan W. Social Filter Theory—A Sleeping Beauty of Regional Innovation Theories. Regional Science and Environmental Economics. 2025; 2(1):2. https://doi.org/10.3390/rsee2010002

Chicago/Turabian Style

Ren, Jianhui, Linlin Lai, Binjie Pei, and Wenyu Zhan. 2025. "Social Filter Theory—A Sleeping Beauty of Regional Innovation Theories" Regional Science and Environmental Economics 2, no. 1: 2. https://doi.org/10.3390/rsee2010002

APA Style

Ren, J., Lai, L., Pei, B., & Zhan, W. (2025). Social Filter Theory—A Sleeping Beauty of Regional Innovation Theories. Regional Science and Environmental Economics, 2(1), 2. https://doi.org/10.3390/rsee2010002

Article Metrics

Back to TopTop