Air Pollution and the Innovation Gap: A Challenge for Sustainable Growth in Emerging and Growth Leading Economies (EAGLE)

Abstract

As emerging economies play a pivotal role in global growth, understanding the impact of environmental factors, especially air pollution, on innovation is crucial. This study empirically examines the relationship between air pollution and innovation performance in EAGLE (emerging and growth leading economies) using a 20-year dataset (2000–2019) and employing 2SLS (two-stage least square) and FMOLS (fully modified ordinary least square) estimation techniques. The key findings of the study reveal that air pollution hampers R&D activities and patent applications by weakening human capital, diverting resources, and creating an unfavorable research environment. The results remain robust across multiple control variables and alternative estimation techniques. These findings highlight the urgent need for pollution control policies to mitigate its adverse effects on innovation, offering valuable insights for policymakers striving to enhance innovation resilience amid environmental challenges. This study uniquely explores the causal impact of air pollution on the macro-level innovation performance of a country.

1. Introduction

In 2019 alone, air pollution was estimated to cost the global economy over $8.1 trillion, equivalent to 6.1% of the global GDP, due to its detrimental effects on health, productivity, and economic growth. Emerging economies, which are often the engines of global innovation, bear a disproportionate burden [1]. For instance, in China, one of the leading innovation hubs, research suggests that a 10% increase in air pollution can reduce R&D investment by nearly 5% as firms divert resources toward healthcare costs and regulatory compliance. Similarly, studies from India indicate that exposure to high pollution levels reduces cognitive performance, directly affecting the creativity and problem-solving abilities of researchers and engineers. These alarming trends underscore that air pollution is not just an environmental crisis but a silent inhibitor of innovation, demanding urgent policy interventions.

The emanating air pollution has created apparent threats to environmental quality, human health, and other lives on the earth [2]. The adverse effects of air pollution have been observed in the shape of respiratory and cardiac diseases and high mortality rates due to direct damage to overall human health [3]. In addition to adverse effects on human health, the increasing air pollution leads to enhanced economic losses in the shape of more expenses for environmental sustainability, an increase in healthcare costs, and exploitation and depletion of natural resources [4]. Looking through the lens of psychological effects, it has been witnessed that increasing air pollution has enhanced the tendency of suicidal attempts, created anxiety and frustration among inhabitants, and significantly reduced the level of happiness [5]. The combined negative effects of air pollution have created great concerns for policy officials taking immediate action against pollution emissions. They are more thrilled for devising such policies that mitigate pollution emissions. Particularly, air pollution has damaged the individual cognitive skills necessary for making judgments and important business decisions [6]. Leaning on arguments, it can be conjectured that the increasing air pollution may have a similar impact on the overall innovation performance of countries. The current study primarily aimed to explore the empirical nexus between air pollution and innovation performance.

A relevant body of literature has highlighted the negative impact of air pollution on many corporate-level decisions. For instance, He, et al., [7] examined the adverse effect of both AP1 and AP2 on the productivity of labor working in multiple industrial towns in China. Chan, [8] conjectured that increasing air pollution damages the efficiency of various business decisions, e.g., cash management and innovation decisions. However, this adverse effect on business decisions can be reverted by focusing on regulatory reforms. Fu, et al., [9] observed the negative impact of air pollution on the production capacity of manufacturing enterprises. He and Lin, [10] indicated that severe air pollution impairs the investment efficiency of enterprises through a deteriorating effect on the health of investment managers. Tan, et al., [11] reveal that an increase in air pollution augments the costs of debt financing because it enhances the default risk of enterprises, which further encourages the lending institution to advance the loans at high rates. Recently, Farooq, et al., [12] quantified the impact of air pollution on corporate investment in BRICS economies and found a negative relationship between air pollution and corporate investment. At a macro level, Dong, et al., [6] asserted that air pollution has an adverse impact on the economic growth of China, and this effect is stronger in the eastern region of China. Such a decrease in economic growth can further impair the efforts (including subsidies, direct benefits, and allocation of funds for innovation) of the central government for more innovation.

Air pollution typically refers to the presence of harmful substances in the air, often resulting from human activities such as industrial processes, vehicular emissions, and the burning of fossil fuels. These pollutants can encompass a range of substances, including particulate matter (PM), carbon monoxide (CO), volatile organic compounds (VOCs), and other hazardous gases and chemicals [13]. These pollutants, when present in the atmosphere in excessive quantities, can have detrimental effects on human health, ecosystems, and the environment. On the flip side, innovation refers to the process of introducing new ideas, methods, products, or processes that create value or improve existing systems. It involves the application of creativity, research, and development efforts to generate novel solutions, technologies, or practices that enhance efficiency, productivity, or quality across various domains [14].

EAGLE countries are characterized by rapid economic expansion, industrialization, and a growing influence on global markets. These economies, which include China, India, Brazil, Indonesia, etc., exhibit high potential for sustained growth but also face significant environmental challenges due to industrial emissions, urban congestion, and weak regulatory enforcement. Their vulnerability to air pollution’s impact on innovation stems from their reliance on manufacturing and energy-intensive industries, which contribute to deteriorating air quality and, in turn, hinder knowledge-based economic activities. When air pollution stifles innovation by reducing worker efficiency, increasing health-related costs, and discouraging investment in high-tech sectors, it poses a substantial threat to long-term economic growth and sustainable development in these economies [15].

By utilizing the country-level data of innovation-related variables from 14 EAGLE (emerging and growth leading economies), we found that pollution emission has a negative impact on all proxies of innovation, including research and development expenditures, patent applications, and trademark applications. The possible mechanism for a negative impact of air pollution on innovation is a reduction in economic growth and government income, which further substantially reduces the allocation of resources for innovation activities. The negative impact of pollution emission on innovation was found to be consistent even by including a series of control variables, including financial development, FDI inflow, personal remittances, real interest rate, and subsidies, and estimating the coefficients in the long run (by employing the FMOLS model). This study contributes to the literature because it explores the empirical relationship between air pollution and country-level innovation activities. Most existing studies only comment on air pollution’s effect on corporate-level innovation activities [16,17,18]. Remarkably, the empirical analysis offers robustness to the existing literature by highlighting the similar effect of air pollution on overall innovation as air pollution showered on firm-level or micro-level innovation performance. This study proposes a policy regarding the adequate control of pollution emission because it has a detrimental effect on the overall innovation performance of a country.

The theoretical significance of examining the relationship between air pollution and innovation lies in its ability to enhance our understanding of how environmental degradation affects economic and technological progress through multiple interconnected channels. Drawing on the endogenous growth theory, this relationship highlights how pollution undermines human capital by impairing health and cognitive function, thereby reducing labor productivity and research efficiency, both essential inputs for innovation-driven growth. The Porter Hypothesis suggests that while pollution typically hinders innovation, well-designed environmental regulations can counteract this effect by incentivizing firms to innovate and adopt cleaner technologies, thus turning regulatory pressure into a competitive advantage. Additionally, the Institutional Theory emphasizes the role of governance structures, where weak environmental institutions in emerging economies fail to enforce compliance or support sustainable R&D, amplifying the negative effects of pollution. Integrating these theoretical frameworks reveals a multi-channel mechanism through which air pollution impacts innovation by directly affecting human capital, distorting resource allocation, weakening institutional support, and altering regulatory incentives. This enriched understanding provides a foundation for refining existing economic models and developing more holistic frameworks that capture the complex interplay between environmental stressors and innovation dynamics.

From an empirical standpoint, investigating the impact of air pollution on innovation holds significant value in providing empirical evidence to support or challenge existing theories and hypotheses. The robust empirical findings derived from a comprehensive dataset spanning different economies and data periods offer tangible evidence of how varying levels of air pollution influence innovation outputs. These empirical insights not only validate or refine theoretical frameworks but also inform policymakers and businesses about the real-world implications of environmental degradation on innovation. This empirical research provides actionable data that can guide the formulation of targeted policies aimed at mitigating pollution’s adverse effects on innovation.

The practical significance of understanding the relationship between air pollution and innovation is substantial in guiding policy interventions and business strategies. The empirical insights derived from the current study can inform policymakers about the urgency of implementing effective pollution control measures to safeguard and foster innovation. Additionally, businesses can leverage this knowledge to strategize and allocate resources towards innovative practices that combat the detrimental effects of pollution. Understanding how pollution affects innovation can aid in the development and adoption of eco-friendly technologies, incentivizing industries to invest in cleaner production methods and sustainable innovations. This practical application of research findings contributes to shaping policies and business practices that prioritize both environmental sustainability and innovation, fostering a more conducive environment for technological advancement while addressing pressing environmental concerns.

This study is structured across several sections, each serving a distinct purpose. Section 2 encompasses the literature review and hypothesis development, offering a comprehensive understanding of existing research and outlining the guiding hypotheses. In Section 3, the focus is on detailing the methodology, encompassing data collection methods and the analytical approaches utilized, ensuring transparency in the study’s methodology. Following this, Section 4 unveils and interprets the empirical findings derived from the applied methodologies. Moving forward, Section 5 engages in a thorough discussion of the study’s findings. Finally, Section 6 encapsulates the study’s conclusion, summarizing key findings and emphasizing the broader significance of the research. This section also integrates a policy discussion, highlighting practical implications and potential interventions stemming from the study’s findings on the relationship between air pollution and innovation.

2. Literature Review and Hypotheses Development

The intersection of air pollution and innovation marks a critical domain in understanding the multifaceted repercussions on a nation’s creative and developmental capacities. Therefore, exploring the nexus between environmental quality and innovation stands pivotal in deciphering the nuanced dynamics shaping a country’s inventive landscape. In this essence, a plethora of literature delves into examining the correlation between air pollution and the advancement of innovation, reflecting a substantial body of research on this intricate relationship. For instance, Ren, et al., [19] aimed to investigate the impact of China’s SO₂ emissions trading program on patent applications, especially environmental patents. They analyzed the data of publicly traded firms from 2004 to 2015 and utilized the DID model to ascertain the regression. Their analysis reveals a substantial uptick among regulated entities. Particulary, an enhanced innovation, notably in strongly enforced environmental zones, correlates with decreased SO2 emissions and augmented industrial development. By utilizing a unique approach involving the central heating policy, the study of Zhang and Chung [14] delves into the relationship between air quality and regional innovation in highly polluted Chinese cities, revealing a substantial, adverse influence of air pollution on city-level innovation. By employing an instrumental variable to address potential biases, the research finds compelling evidence indicating a detrimental impact of air pollution on innovation, unlinking this effect from skilled worker self-selection and attributing it to productivity decline.

Mehmood, et al., [20] examined how financial development and environmental degradation influence the control of corruption using system GMM on annual data from 90 countries. The findings revealed that financial development weakens corruption control, while environmental degradation strengthens it, highlighting policy implications for financial and environmental governance.

In a pioneering study, Liu, et al., [21] aimed to find out the relationship between air pollution and research and development (R&D) in China from 2007 to 2016. The findings of their study reveal a notable detrimental effect of air pollution on R&D input and output. Employing diverse indicators and instrumental variable methods, the study further highlights a direct correlation: a 1% increase in PM2.5 leads to a reduction of 0.359% in annual R&D personnel, 0.169% in expenditures, and 0.293% in new patents, signifying air pollution’s substantial hindrance to innovation and technological advancement in the region. Wang and Wu [22] utilized a partial equilibrium model to unveil the impact of air pollution on the exodus of high-skilled professionals in China and India. The findings of their study highlight a stark decline in technological innovative human capital due to air pollution, with a 1% rise in PM2.5 concentration leading to a notable drop of 146 professionals in China and a 0.127% decrease in TIP stock in Indian states for every 1% increase in PM10 concentration. In addition, regional nuances reveal a pronounced negative influence in affluent Chinese areas, while underdeveloped regions seem less affected, underlining the urgency for targeted interventions to mitigate brain drain caused by air pollution. Wang, et al., [23] investigated the impact of air pollution on innovation among Chinese listed firms, revealing a marked decline in innovation within highly polluted city headquarters. Employing a 2SLS method, the study establishes a strong causal link using local thermal inversions as instrumental variables, uncovering how this decline is driven by brain drain, particularly the migration of skilled workers, especially impactful in lower-paying firms and competitive industries.

Similarly, Tan and Yan [24] unveiled a link between urban air pollution and corporate innovation. They vowed that increased air pollution diminished corporate innovation. The analysis confirms this relationship empirically, demonstrating that heightened pollution levels correlate with heightened financial constraints on firms and a decline in human capital, impacting innovation investments and talent acquisition. Zhu and Lee [25] employed a spatial dynamic panel data model, analyzing PM2.5’s impact on innovation from 2001 to 2016. Using real-time data from WDI on the innovation index, the findings of their study reveal a pronounced hindrance of PM2.5 on innovation at both regional and neighboring levels, persisting even after addressing endogeneity concerns. Lin, et al., [26] delved into how air pollution affects technological innovation by constructing models and analyzing data from China’s provinces between 2004 and 2017. The findings reveal that higher air pollution levels impede innovation by crowding out investment, particularly affecting invention and utility model patents, suggesting the need for targeted environmental policies to improve air quality and optimize innovation fund allocation.

Ai, et al., [13] investigated the relationship between air pollution and urban innovation by utilizing a fixed-effect model across 281 Chinese prefecture-level cities from 2003 to 2015. The analysis reveals a significant hindrance by air pollution on overall urban innovation. Specifically, larger-scale and eastern cities, alongside key environmental protection cities, experience amplified negative effects, indicating a nuanced impact of air pollution on urban innovation capability. Bu and Zhang [27] investigated the impact of air pollution on Chinese industrial firms’ innovation, utilizing the Qinling Mountains-Huai River heating policy as a quasi-natural experiment. The analysis reveals a substantial negative correlation, with a 1% rise in PM2.5 emissions resulting in significant reductions in corporate total innovation, invention patents, utility model patents, and design patents, highlighting air pollution’s inhibitory effect, particularly on green innovation, state-owned enterprises, high pollution, and low-monopoly firms. While examining Chinese A-share-listed companies’ data from 2004 to 2017, Sun, et al., [28] employed a difference-in-differences approach, demonstrating that air pollution collaborative governance significantly fosters green technology innovation. Their analysis further indicates that while this governance boosts green utility model patents, its impact on invention patents is negligible. Furthermore, it primarily spurs innovation among non-heavy-polluting enterprises, with R&D investment mediating its positive influence and government subsidies augmenting its efficacy.

Zhu, et al., [29] utilized China’s provincial data to reveal how air pollution negatively impacts regional innovation, employing advanced statistical models to address endogeneity issues. The findings of their study indicate that a 1% rise in PM2.5 leads to a 0.186% decrease in patent applications, with robust results across various sample classifications. Additionally, the research underscores the pivotal role of R&D personnel’s health, suggesting that reducing pollution and mitigating work pressure can enhance regional innovation alongside sustained investment in innovation capital. Examining data from Chinese listed firms between 2007 and 2019, Ma and He [30] suggested a significant positive correlation between ambient air pollution at firm headquarters and subsequent green innovation efforts. This positive relationship, supporting stakeholder theory, is amplified in firms with higher analyst coverage, corporate site visits, and institutional ownership. Furthermore, the study demonstrates the robustness of this association through exogenous factors such as strong wind and rainfall and highlights the positive impact of revisions in environmental protection laws on fostering the link between air quality and corporate green innovation.

Recently, Song, et al., [31] employed regression discontinuity design to uncover the distinct link between air pollution and urban innovation in China, revealing a notable 1.463% decrease in urban innovation for every 1% rise in air pollution. Moreover, their study highlights that air pollution disproportionately affects invention patents, particularly those demanding higher innovation capabilities, suggesting a potential mechanism involving compromised human capital and increased migration of skilled individuals from urban areas. Xu, et al., [32] examined the impact of air pollution on technology transfer, using thermal inversion as an instrumental variable to address endogeneity. The analysis reveals a 4.5% reduction in technology transfer with increased pollution, uncovering a nuanced asymmetry: highly polluted cities seek technology transfer to other provinces, with varying effects based on GDP levels, highlighting divergent impacts and the role of environmental regulations.

The existing literature rigorously explores the impact of air pollution on innovation across diverse sectors, specific industry clusters or sectors, and geographic scales, unveiling intricate relationships and specific causal mechanisms. However, despite the comprehensive nature of these studies, there remains a notable gap in understanding the nuanced impact of air pollution on country-level innovation performance, thereby requiring targeted investigations into how pollution uniquely influences innovation dynamics within specialized economic domains. This gap could shed light on tailored intervention strategies to mitigate pollution’s adverse effects on innovation within these economies and foster country-specific innovation growth strategies. To address the existing gap in the literature, the following hypotheses can be developed:

H1: 

Air pollution has a significant negative relationship with research and development activities.

H2: 

Air pollution has a significant negative relationship with patent applications.

3. Data and Methods

3.1. Data and Sample

The empirical analysis utilized 20 years of data spanning from 2000 to 2019, focusing on 14 EAGLE (emerging and growth leading) economies, with specific country details provided in Appendix A 

Table A1. Average Values Across the Sample Countries.

Sr. no.	Country	RDE	TDA	PTA	CO2	AP1	AP2	FDI	PRM	FSD	RIR	SUB
1	Brazil	1.122	121,814.313	4393.400	2.006	14.850	84.490	3.269	0.214	48.161	36.723	57.064
2	China	1.624	2,364,147.188	490,844.350	5.659	61.006	100.000	3.079	0.200	128.813	2.042	50.091
3	Egypt	0.490	12,285.313	652.684	2.149	76.967	100.000	2.930	5.734	38.817	2.618	37.442
4	India	0.742	180,116.063	8577.250	1.304	90.288	100.000	1.631	3.176	44.650	5.125	38.245
5	Indonesia	0.122	45,377.143	735.526	1.717	17.587	97.173	1.270	1.011	29.422	4.948	59.342
6	Iran	0.503	324,660.250	8731.474	6.797	38.318	100.000	0.862	0.558	43.458	0.328	31.368
7	Korea, Rep.	3.288	158,208.563	130,772.300	10.999	28.843	100.000	0.961	0.568	124.777	3.502	58.785
8	Malaysia	0.945	14,768.750	858.950	6.798	18.063	95.083	3.201	0.528	115.517	2.339	41.526
9	Mexico	0.392	74,398.063	922.600	3.977	25.219	99.861	2.771	2.236	23.869	1.897	57.091
10	Pakistan	0.341	17,878.933	138.211	0.742	61.881	100.000	1.178	4.958	20.308	1.860	3.223
11	Philippines	0.143	20,911.100	245.500	0.963	21.515	99.382	1.603	10.457	34.692	4.247	23.166
12	Russian Fed.	1.104	152,720.125	25,801.100	11.251	18.156	94.623	2.211	0.384	39.899	−0.222	59.084
13	Turkey	0.735	172,312.744	3311.300	4.025	42.596	100.000	1.596	0.376	41.484	4.012	49.508
14	Vietnam	0.413	46,787.100	320.550	1.664	35.729	100.000	4.931	5.364	76.619	1.626	48.091

Acronyms: RDEs = research and development expenditures, TDAs = trademark applications, PTAs = patent applications, CO₂ = CO₂ emissions, AP1= pollution emissions 1, AP2 = pollution emissions 2, FDI = foreign direct investment inflow, PRMs = personal remittances, FSD = financial sector development, RIR = real interest rate, SUB = subsidies Source: self-elaboration. Note: This Table shows the average values of all variables across the sample nations.

. Initially, the sample comprised 16 countries outlined in BBVA (Banco Bilbao Vizcaya Argentaria) research. However, Bangladesh and Nigeria were excluded due to lacking information on research and development expenditures (RDEs) and patent applications (PTAs). The selection of the remaining 14 EAGLE economies for the analysis stems from their status as emerging economies expected to significantly contribute to global economic growth, surpassing the average contributions of Western countries, notably the U.S. and the European Union, as highlighted by BBVA [33]. These economies, experiencing rapid economic expansion, are likely to encounter heightened environmental challenges, such as air pollution, impacting various decisions, including innovation. The study’s focus on this specific set of economies is driven by the expectation that their rapid economic expansion may lead to increased environmental challenges, notably heightened air pollution, potentially influencing decision-making processes, particularly in the realm of innovation. Thus, focusing the analysis on EAGLE economies promises to offer deeper insights into the interplay between environmental issues and innovation.

Similarly, the selection of this timeframe aligns with the need to observe trends and patterns over a substantial period, allowing for a comprehensive assessment of how these economies have navigated environmental challenges while striving for innovation-driven growth. The span from 2000 to 2019 encompasses significant global economic shifts and environmental changes, offering a robust temporal scope to study the interconnected dynamics between economic emergence, environmental concerns, and innovation outputs within these economies. Data for the analysis were sourced from WDI, The World Bank.

The EAGLE economies provide a particularly compelling context for this research due to their rapid economic growth, industrial expansion, and increasing environmental challenges. These economies are key drivers of global innovation, yet they also face significant air pollution concerns that could hinder their technological progress. Given their strategic position in global supply chains and their substantial contributions to global GDP, understanding how environmental factors influence innovation in these economies is crucial. Neglecting this topic could have severe economic and social consequences, as deteriorating air quality can reduce labor productivity, increase healthcare costs, and weaken human capital formation, ultimately stalling long-term innovation and competitiveness. Moreover, failing to address these challenges may exacerbate economic inequality and social unrest, as vulnerable populations bear the brunt of pollution-related health and economic burdens. Therefore, this study is essential in highlighting the urgent need for policies that balance economic growth with sustainable innovation.

While the dataset used in this study is comprehensive, covering 20 years of data across EAGLE economies, certain limitations and potential biases should be acknowledged. First, data availability constraints may have led to the exclusion of some relevant variables or countries, potentially affecting the generalizability of findings. Second, while we employ multiple proxies for air pollution and innovation, measurement errors or discrepancies in data reporting across countries may introduce inconsistencies. Additionally, unobserved factors, such as institutional quality or informal innovation activities, might influence the results but are challenging to quantify. Despite these limitations, robust econometric techniques have been applied to mitigate biases and enhance the reliability of the findings.

3.2. Research Models

The analysis is based upon multiple linear regression models, which can be structured as follows:

$${RDE}_{it}={\beta }_{°}+{\alpha }_1{CO2}_{it}+{\alpha }_2{AP1}_{it}+{\alpha }_3{AP2}_{it}+{\beta }_1{FDI}_{it}+{\beta }_2{PRM}_{it}+{\beta }_3{FSD}_{it}+{\beta }_4{RIR}_{it}+{\beta }_5{SUB}_{it}+{\epsilon }_{it}$$

(1)

$${TDA}_{it}={\beta }_{°}+{\alpha }_1{CO2}_{it}+{\alpha }_2{AP1}_{it}+{\alpha }_3{AP2}_{it}+{\beta }_1{FDI}_{it}+{\beta }_2{PRM}_{it}+{\beta }_3{FSD}_{it}+{\beta }_4{RIR}_{it}+{\beta }_5{SUB}_{it}+{\epsilon }_{it}$$

(2)

$${PTA}_{it}={\beta }_{°}+{\alpha }_1{CO2}_{it}+{\alpha }_2{AP1}_{it}+{\alpha }_3{AP2}_{it}+{\beta }_1{FDI}_{it}+{\beta }_2{PRM}_{it}+{\beta }_3{FSD}_{it}+{\beta }_4{RIR}_{it}+{\beta }_5{SUB}_{it}+{\epsilon }_{it}$$

(3)

Equation (1) primarily demonstrates the impact of air pollution proxies, including CO₂ (CO₂ emissions), AP1 (air pollution 1), and AP2 (air pollution 2) on RDE (research and development) of country $i$ in time t. The control variables are FDI (FDI inflow), PRMs (personal remittances), FSD (financial sector development), RIR (real interest rate), and SUB (subsidies). Equation (2) shows the effect of air pollution on TDAs (trademark applications), and Equation (3) is for showing the effect of air pollution on PTAs (patent applications).

Table 1. Description of Variables.

Acronym	Variables	Measurement	Role	Reference	Data Source
RDE	Research and development	Research and development expenditures (% of GDP)	Dependent	[18,34,35]	WDI, The World Bank
TDAs	Trademark applications	Trademark applications, resident, by count	Dependent	[36,37]	WDI, The World Bank
PTAs	Patent applications	Patent applications, residents	Dependent	[18,36,38]	WDI, The World Bank
CO2	CO2 emissions	CO2 emissions (metric tons per capita)	Independent	[12,17]	WDI, The World Bank
AP1	Air Pollution 1	PM2.5 air pollution, mean annual exposure (micrograms per cubic meter)	Independent	[12,39]	WDI, The World Bank
AP2	Air Pollution 2	PM2.5 air pollution, population exposed to levels exceeding WHO guideline value (% of total)	Independent	[12,39]	WDI, The World Bank
FDI	Foreign Direct Investment	Foreign direct investment, net inflows (% of GDP)	Control	[40]	WDI, The World Bank
PRMs	Personal remittances	Personal remittances, received (% of GDP)	Control	[41]	WDI, The World Bank
FSD	Financial sector development	Domestic credit to the private sector (% of GDP)	Control	[42]	WDI, The World Bank
RIR	Real interest rate	Real interest rate (%)	Control	[43]	WDI, The World Bank
SUB	Subsidies	Subsidies and other transfers (% of expense)	Control	[23]	WDI, The World Bank

Source: self-elaboration. Note: This table shows the measurement of variables.

shows the brief measurement of variables.

The subscript it denotes country i in year t, where t = 2000, …, 2019, indicating the use of yearly panel data. AP1 refers to the mean annual exposure to PM2.5 air pollution, measured in micrograms per cubic meter, while AP2 captures the percentage of the population exposed to PM2.5 levels that exceed the WHO guideline value. Including both indicators allows us to assess air pollution’s impact on innovation from two complementary dimensions: AP1 reflects the overall intensity of pollution exposure in the environment, whereas AP2 highlights the extent of population vulnerability to harmful pollution levels. Although both relate to PM2.5, they are not redundant; AP1 quantifies concentration levels, while AP2 emphasizes the scale of public exposure, enabling a more nuanced understanding of how air pollution affects innovation performance across different socio-environmental contexts.

The inclusion of FDI and subsidies as control variables is particularly relevant in the context of EAGLE economies, given their significant role in shaping innovation dynamics. FDI serves as a critical channel for technology transfer, knowledge spillovers, and R&D collaborations, making it an essential determinant of innovation performance in these rapidly growing markets. Higher FDI inflows can mitigate the negative effects of air pollution by providing financial and technological resources that enhance firms’ adaptive capacity. Similarly, government subsidies play a vital role in fostering innovation by offsetting the costs associated with R&D, particularly in environments where pollution-induced uncertainties may discourage private sector investment in innovation. Given that EAGLE economies are characterized by high economic growth, industrialization, and varying levels of policy support, these controls ensure a more accurate estimation of the relationship between air pollution and innovation performance. While education and institutional quality are important, they were excluded due to multicollinearity concerns.

3.3. Methodology

It is expected that air pollution and innovation performance are endogenous to each other; therefore, we choose the 2SLS (two-stage least squares) model for regression analysis and check the robustness through the FMOLS (fully modified ordinary least square) model. The selection of both models is purely based upon some pre-estimation techniques. As the analysis was conducted on several countries, the probability of cross-section dependence (CD) is high. To investigate the existence of CD, we develop Equation (4) and employ the family of CD techniques suggested by Breusch and Pagan [44]. Out of three, two techniques, i.e., Breusch-Pagan LM and Breusch scaled LM, have significant probability values, rejecting the null hypothesis, i.e., no cross-section dependence among series. In the presence of CD, we further test the stationarity of the series by employing the second-generation unit root test suggested by Pesaran [45].

The reported values of all variables speak about the stationarity of series at level 1 I(1), which further suggests checking the cointegration among series. For this, we develop Equation (5) and test it by employing the Johansen cointegration test [46] and select the Kao-residual technique for testing the cointegration. The analysis confirms the existence of cointegration and argues to estimate the coefficients in the long run by employing the FMOLS model. The employability of 2SLS as a baseline model is based upon the endogenous nature of both variables, i.e., air pollution and innovation. Both variables can determine each other simultaneously and therefore can be regarded as endogenous variables. Wang, et al., [22], and Zhao, et al., [47] employed the 2SLS model for conducting a similar analysis at the micro-level. In addition, the causality among variables was tested by employing the pairwise Granger causality test [48].

The choice of 2SLS over other econometric methods such as GMM is primarily driven by concerns of endogeneity in the relationship between air pollution and innovation performance. Air pollution can be influenced by economic activity, creating a potential reverse causality issue where higher innovation levels might lead to increased industrial emissions. Thereby, 2SLS addresses this by using instrumental variables (IVs) that are correlated with air pollution but not directly with innovation, ensuring a more robust estimation of causal effects. While GMM is a powerful tool for addressing endogeneity, it is more suited for dynamic panel data models with short time spans and requires strict assumptions on instrument validity. Given the study’s 20-year dataset and focus on long-term relationships, 2SLS provides a more reliable approach for mitigating endogeneity concerns while ensuring the consistency of estimates.

In the 2SLS model, we employ lagged values of the air pollution variables as instruments to address potential endogeneity concerns. The theoretical justification for using lagged air pollution measures lies in their strong temporal correlation with current pollution levels due to the persistent nature of environmental degradation, driven by structural industrial and urban dynamics. However, these lagged values are unlikely to have a direct, contemporaneous impact on innovation outcomes such as R&D expenditures or patent applications. Innovation decisions typically respond to current macroeconomic and environmental conditions rather than past pollution exposure, particularly given the relatively short-term planning horizons of many innovation investments in emerging economies. Thus, the lagged pollution variables fulfill the exclusion restriction by being strongly correlated with current pollution (the endogenous regressor) but plausibly exogenous to current innovation performance, making them valid instruments for identification in the 2SLS framework. This inclusion of lag variables as an instrument was also supported by Cui, et al., [49], and Hao, et al., [50].

$$CD=\sqrt{{\displaystyle \frac{2T}{N(N-1)}} \left({\sum }_{i=0}^{N-1}{\sum }_{j=i+1}^N\rho ij\right)}$$

(4)

$${\rho }_{it}={\mu }_i{\rho }_{it-1}+{\hat{\gamma }}_{it}$$

(5)

4. Empirical Results

4.1. Descriptive and Correlation Analysis

Table 2. Descriptive Analysis.

	Mean	Median	Maximum	Minimum	Std. Dev.	Skewness	Kurtosis
RDE	1.075	0.804	4.627	0.084	1.031	1.979	6.254
TDA	93,984.040	82,761.000	322,297.000	4340.000	74,841.370	0.673	2.721
PTA	23,465.980	3956.000	171,603.000	162,000	46,771.310	2.243	6.445
CO2	5.128	3.591	12.225	0.823	4.018	0.656	1.800
AP1	36.143	22.133	97.599	12.658	28.286	1.191	2.639
AP2	96.209	99.715	100.000	67.952	6.333	−2.602	10.401
FDI	2.409	2.301	8.876	−0.204	1.349	1.019	6.088
PRM	2.468	0.834	10.493	0.107	2.972	1.500	4.146
FSD	60.957	47.973	151.258	14.776	37.356	0.993	2.530
RIR	6.465	3.397	44.635	−12.856	11.356	1.991	6.298
SUB	49.314	51.195	71.209	22.326	12.773	−0.395	2.175

Abbreviation: RDEs = research and development expenditures, TDAs = trademark applications, PTAs = patent applications, CO₂ = CO₂ emissions, AP1 = pollution emissions 1, AP2 = pollution emissions 2, FDI = foreign direct investment inflow, PRMs = personal remittances, FSD = financial sector development, RIR = real interest rate, SUB = subsidies Source: self-elaboration. Note: This table shows the summary of descriptive statistics.

shows the accumulated descriptive analysis, and Appendix A Table A1 presents the average values of all variables across the sample countries. The mean value of RDE is 1.075%, showing the percentage of research and development expenditures made as compared to total GDP. In individual analysis, the Korean Republic has the highest value of RDE, which is 3.388% of GDP (shown in Appendix A Table A1). Similarly, the overall mean value of CO₂ is 5.128 MTP (metric tons per capita), which is quite higher than the 4MTP of the rest of the world, implying that EAGLE economies have massive pollution emissions.

Table 3. Correlation Analysis.

	RDE	TDA	PTA	C02	AP1	AP2	FDI	PRM	FSD	RIR	SUB
RDE	1.000
TDA	0.428	1.000
PTA	0.943	0.446	1.000
CO2	0.682	0.309	0.674	1.000
AP1	−0.155	0.126	−0.107	−0.406	1.000
AP2	0.008	−0.152	0.202	0.072	0.437	1.000
FDI	−0.352	−0.303	−0.443	−0.181	−0.082	−0.290	1.000
PRM	−0.432	−0.397	−0.307	−0.542	0.417	0.382	−0.026	1.000
FSD	0.781	0.166	0.687	0.591	−0.237	−0.027	−0.156	−0.438	1.000
RIR	0.022	0.174	−0.125	−0.327	−0.251	−0.733	0.104	−0.273	−0.053	1.000
SUB	0.412	0.459	0.382	0.511	−0.438	−0.273	−0.039	−0.641	0.120	0.198	1.000
Multicollinearity Analysis-Uncentered VIF (variance inflation factor)
VIF	1.998	2.091	1.971	1.307	2.100	1.682	1.210	1.036	2.440	1.237	1.231

Abbreviation: Please see the list of abbreviations. Source: self-elaboration. Note: This table shows the statistics of correlation analysis.

demonstrates the correlation values among the variables. We also check the multicollinearity by estimating the VIF (variance inflation factor) values of variables (presented in the bottom lines of Table 3). In addition, we show the emissions trend of countries in Figure 1. The figure illustrates the CO₂ emissions trends across selected EAGLE economies from 2000 to 2019. The data reveals that China and Iran exhibit the highest levels of emissions, with a noticeable upward trajectory over the years. India and Russia also show significant increases, reflecting their industrial expansion and economic growth. In contrast, countries such as Brazil, Indonesia, and the Philippines maintain relatively lower emissions, although they exhibit a gradual rise. The trends suggest that while some economies experience fluctuations, overall CO₂ emissions have been increasing across most sample countries, indicating the pressing need for sustainable policies to curb pollution and mitigate its environmental and economic impact.

4.2. Pre-Estimation Analysis

Before regression analysis, we check the cross-section dependence, stationarity among series, and cointegration and report the analyses in

Table 4. Testing Cross-Section Dependence.

Test	Statistics	D.F	Probability
Breusch-Pagan LM	126.101	45	0.000
Pesaran scaled LM	7.494	-	0.000
Pesaran CD	0.260	-	0.794

Note: The significant values accept the alternative hypothesis, i.e., there is an issue of cross-section dependence among the series. Source: self-elaboration.

,

Table 5. Unit Root Testing.

	(CIPS)		(CADF)
Variables	At Level	At First Difference	At Level	At First Difference
RDE	(2.706) 0.996	(−4.418) 0.000 ***	(16.834) 0.951	(74.647) 0.000 ***
TDA	(5.845) 0.990	(−2.704) 0.003 ***	(3.092) 0.998	(51.473) 0.028 ***
PTA	(6.068) 0.997	(−5.934) 0.000 ***	(9.050) 0.999	(92.576) 0.000 ***
CO2	(−1.664) 0.952	(−4.360) 0.000 ***	(22.802) 0.742	(67.408) 0.000 ***
AP1	(2.184) 0.985	(−4.922) 0.000 ***	(11.648) 0.997	(73.914) 0.001 ***
AP2	(0.954) 0.830	(−5.097) 0.000 ***	(6.928) 0.937	(54.374) 0.000 ***
FDI	(−3.681) 0.135	(−69.402) 0.000 ***	(−4.269) 0.159	(86.368) 0.000 ***
PRM	(−5.711) 0.198	(−4.172) 0.000 ***	(71.552) 0.374	(68.551) 0.000 ***
FSD	(1.203) 0.885	(−3.429) 0.000 ***	(18.113) 0.923	(56.133) 0.000***
RIR	(−3.228) 0.771	(−3.799) 0.000 ***	(58.011) 0.671	(95.366) 0.000 ***
SUB	(−0.938) 0.173	(−7.313) 0.000 ***	(27.891) 0.178	(91.919) 0.000 ***

Acronyms: Please see the list of abbreviations. Source: self-elaboration. Note: The reported values infer that all variables are stationary at level 1 I(1). *** is the significance levels at 1%.

, and

Table 6. Cointegration Analysis.

Kao Residual Cointegration Technique
Test Name	t-Statistics	Probability
ADF	−2.618	0.004
Residual Variance	0.008	-
HAC Variance	0.007	-

Note: The analysis shows that the ADF test has a significant value, authorizing the existence of cointegration. Source: self-elaboration.

, respectively. The underlying analyses report that there exists the issue of cross-section dependence, which further leads to checking the stationarity through second-generation unit root testing. All the variables are stationary at level I(1). Similarly, the statistical analysis shown in Table 6 accepts the alternative hypothesis, i.e., the cointegration exists.

4.3. Main Regression Analysis

Table 7. Effect of Pollution Emissions on Innovation.

	Panel Two-Stage Least Square (2SLS)
	RDE as a Dependent		TDA as a Dependent		PTA As a Dependent
Variables	Coefficients	Probability	Coefficients	Probability	Coefficients	Probability
C	−5.416 ***	0.000	0.343 **	0.065	−0.285 ***	0.000
CO2	−0.153 ***	0.000	−0.008 **	0.069	−0.606 ***	0.000
AP1	−0.010 ***	0.000	−0.164 ***	0.000	−0.267 ***	0.001
AP2	−0.030 ***	0.006	−0.351 **	0.051	−0.191 ***	0.002
FDI	0.174 ***	0.000	0.302 ***	0.000	0.176 ***	0.000
PRM	0.103 ***	0.000	−0.460	0.171	0.498 ***	0.000
FSD	0.015 ***	0.000	−0.168	0.472	0.580 ***	0.000
RIR	−0.050 ***	0.000	0.146 ***	0.248	−0.179 ***	0.000
SUB	0.025 ***	0.000	0.175 ***	0.046	0.129 ***	0.000
Adjusted R-square		0.691	0.408		0.682
S.E. of regression		0.444	0.583		0.235
Probability (F-statistics)		0.000	0.000		0.000
Second-Stage SSR		15.418	2.422		2.900

Acronyms: RDEs = research and development expenditures, TDAs = trademark applications, PTAs = patent applications, CO₂ = CO₂ emissions, AP1 = pollution emissions 1, AP2 = pollution emissions 2, FDI = foreign direct investment inflow, PRMs = personal remittances, FSD = financial sector development, RIR = real interest rate, SUB = subsidies Source: self-elaboration. Note: *** and ** are the significance levels at 1% and 5% respectively. Instruments: RDE (−1), TDA (−1), PTA (−1), CO₂ (−1), AP1 (−1), AP2 (−1), FDI (−1), PRM (−1), FSD (−1), RIR (−1), and SUB (−1).

displays the main regression analysis for the panel 2SLS model. All proxies of pollution emission, including CO₂, AP1, and AP2, have negative and statistically significant coefficient values, implying that pollution emission has an adverse impact on innovation variables, i.e., RDE, TDA, and PTA. For control variables, FDI inflow, PRM, FSD, and SUB show a positive relationship with innovation variables, while RIR carries a negative relationship with innovation activities.

The stronger negative impact of CO₂ emissions on patent applications (−0.606) compared to R&D expenditures (−0.153) can be interpreted through the lens of innovation output versus input dynamics. While R&D spending represents the input side of innovation, reflecting ongoing investments in knowledge creation, patent applications signify the output or tangible outcomes of these efforts. High levels of CO₂ emissions are often associated with poor environmental quality, deteriorating health conditions, and lower cognitive productivity, which can severely disrupt the complex and time-sensitive processes required to convert R&D efforts into patentable innovations. In contrast, R&D expenditures may be less immediately sensitive to environmental degradation, particularly if they are tied to long-term institutional budgets or external funding commitments. Moreover, the uncertainty and risks introduced by pollution may lead enterprises and inventors to delay or abandon the patenting process due to lower expected returns, increased compliance burdens, or resource diversion toward pollution mitigation efforts. Thus, the larger negative coefficient for patents reflects the compounded downstream effects of pollution on the final stages of the innovation pipeline.

4.4. Robustness Analysis

The robustness of the findings was checked by testing the coefficients in the long run through the FMOLS model. The analysis reported in

Table 8. Robustness Analysis (Effect of Pollution Emissions on Innovation).

	FMOLS (Fully Modified Ordinary Least Square)
	RDE as a Dependent		TDA as a Dependent		PTA as a Dependent
Variables	Coefficients	Probability	Coefficients	Probability	Coefficients	Probability
CO2	−0.387 ***	0.001	−0.347 ***	0.015	−0.524 ***	0.018
AP1	−0.022 **	0.053	−0.306 **	0.057	−0.335 **	0.078
AP2	−0.058 *	0.096	−0.319 **	0.083	−0.525 ***	0.025
FDI	0.061 ***	0.012	−0.175	0.793	0.108 ***	0.041
PRM	0.051 ***	0.047	0.339 ***	0.020	0.390 **	0.087
FSD	0.001 ***	0.006	0.832	0.451	0.259	0.183
RIR	−0.006 ***	0.043	−0.206 **	0.087	−0.138 **	0.061
SUB	0.024 ***	0.002	0.573 ***	0.001	0.181 ***	0.040
Adjusted R-square		0.401	0.458		0.374
S.E. of regression		0.141	0.571		0.325
Long run variance		0.111	0.007		0.312

Acronyms: Please see the list of abbreviations. Source: self-elaboration. Note: ***, **, and * are the significance levels at 1%, 5%, and 10%, respectively.

confirms the consistency of the relationship as shown in Table 7. We further check the causality among variables and report the analysis in Appendix A 

Table A2. Pairwise Causality Analysis.

Null Hypothesis (H0)	F-Statistic	Prob.
TDA does not Granger Cause RDE	0.288	0.750
RDE does not Granger Cause TDA	0.366	0.693
PTA does not Granger Cause RDE	0.079	0.923
RDE does not Granger Cause PTA	0.804	0.448
CO2 does not Granger Cause RDE	4.312	0.014
RDE does not Granger Cause CO2	0.815	0.443
AP1 does not Granger Cause RDE	0.032	0.967
RDE does not Granger Cause AP1	1.735	0.178
AP2 does not Granger Cause RDE	2.844	0.060
RDE does not Granger Cause AP2	1.051	0.351
FDI does not Granger Cause RDE	0.210	0.810
RDE does not Granger Cause FDI	0.812	0.445
PRM does not Granger Cause RDE	0.015	0.984
RDE does not Granger Cause PRM	0.414	0.661
FSD does not Granger Cause RDE	5.363	0.005
RDE does not Granger Cause FSD	2.012	0.136
RIR does not Granger Cause RDE	0.725	0.485
RDE does not Granger Cause RIR	0.076	0.926
SUB does not Granger Cause RDE	1.341	0.264
RDE does not Granger Cause SUB	2.135	0.121

Acronyms: RDEs = research and development expenditures, TDAs = trademark applications, PTAs = patent applications, CO₂ = CO₂ emissions, AP1= pollution emissions 1, AP2 = pollution emissions 2, FDI= foreign direct investment inflow, PRMs = personal remittances, FSD = financial sector development, RIR = real interest rate, SUB = subsidies Source: self-elaboration. Note: The significant probability values rejects the null hypothesis and indicates the existence of causality among variables. Source: self-elaboration.

. The causality analysis provides valuable insights into the directional relationships between air pollution and innovation performance. The results indicate that CO₂ emissions Granger cause R&D expenditure, suggesting that worsening air quality has a significant predictive impact on research and development activities. However, the reverse causality is not supported, implying that changes in R&D investment do not directly influence CO₂ levels. Similarly, PM2.5 (AP2) shows a marginally significant effect on RDE (p = 0.060), reinforcing the notion that pollution levels negatively affect innovation capacity. Furthermore, FSD Granger causes RDE (p = 0.005), highlighting the crucial role of financial infrastructure in fostering innovation. These findings underscore the unidirectional impact of environmental degradation on innovation, emphasizing the need for stronger pollution control policies and financial incentives to sustain technological progress in EAGLE economies.

5. Discussion on Results

To achieve the aim of the study, we mainly employ the 2SLS model and check the robustness through the FMOLS model. The current analysis reflects that innovation performance, measured through R&D intensity and trademark applications, captures the collective efforts of both governments and industrial actors toward technological advancement. The significant negative relationship between air pollution and innovation, as revealed by our 2SLS and FMOLS estimates, can be theoretically grounded in the endogenous growth theory, which emphasizes the role of human capital and R&D investment in driving innovation-led growth. At the macro level, elevated air pollution constrains economic development [6], which in turn reduces fiscal space for R&D funding, weakening innovation ecosystems. This aligns with the theory that declining public and private investment in knowledge creation undermines long-term growth potential. At the micro level, as innovation becomes financially and ethically unsustainable in polluted environments, firms face resource reallocation and human capital attrition—consistent with the theory of induced innovation, where adverse environmental conditions distort firms’ innovation choices. This, as noted by Boeing et al. [51], may lead to talent outflows and a decline in innovation productivity.

At the micro level, the adverse impact of air pollution on innovation can be interpreted through multiple theoretical lenses. According to endogenous growth theory, human capital and innovation are central to sustained economic progress. However, our findings show that elevated air pollution undermines these drivers by increasing default risks, damaging health, reducing labor efficiency, and impairing cognitive functions—factors that collectively bias innovation investment forecasts and discourage risk-taking behavior among managers. This aligns with Wang et al. [16], Zhao et al. [47], and Chen et al. [18], who observed similar micro-level deterrents to enterprise-level innovation in polluted environments.

Furthermore, air pollution reallocates critical resources away from innovation. As healthcare costs rise and productivity declines, firms divert financial and managerial attention from R&D to pollution-related contingencies, weakening innovation capacity. This supports the endogenous growth framework’s emphasis on the productivity of knowledge capital and highlights how environmental degradation diminishes its returns.

Simultaneously, the Institutional Theory sheds light on how weak environmental governance in emerging economies exacerbates the negative externalities of pollution. In such contexts, the lack of enforcement and incentives for clean innovation discourages firms from investing in sustainable R&D. Additionally, as Song et al. [31] noted, pollution damages the physical infrastructure and research environment essential for innovation, further eroding innovation performance.

From the perspective of the Porter Hypothesis, however, there is potential for a counter-narrative: in well-regulated environments, stringent environmental policies can incentivize firms to innovate and develop cleaner technologies. Yet, in the EAGLE economies analyzed, characterized by uneven regulatory enforcement, this potential is not fully realized, and air pollution primarily serves as a constraint rather than a catalyst. In sum, the cumulative effects of pollution, manifested through reduced human capital, misallocated resources, impaired infrastructure, and weak institutional frameworks, significantly suppress innovation, as reflected in our results on declining R&D intensity and patent applications across EAGLE nations.

Neglecting the impact of air pollution on innovation can have severe long-term consequences, including reduced competitiveness, brain drain, and slowed technological advancement. Poor air quality negatively affects cognitive function and worker productivity, leading to diminished research output and innovation capacity. This can result in declining global competitiveness, as firms in polluted regions struggle to keep pace with counterparts in cleaner environments. Moreover, skilled professionals may migrate to less polluted areas, exacerbating talent shortages and weakening domestic innovation ecosystems. A decline in technological progress hampers industrial transformation, slowing the development of sustainable solutions and clean technologies [51]. On a broader scale, stagnation in innovation impedes global economic growth, as economies reliant on outdated technologies become less efficient and struggle to compete in a rapidly evolving world. Environmentally, reduced progress in green technologies prolongs dependence on fossil fuels and delays climate action, worsening ecological degradation. Addressing air pollution is therefore not just a public health issue but a critical factor in sustaining innovation-driven economies and ensuring long-term environmental and economic sustainability.

The positive effect of FDI inflow on innovation can be explained as the incoming foreign investment promoting favorable competition among enterprises, and therefore they struggle more to update the existing production system. Moreover, FDI inflow brings new technology, modern knowledge, and necessary financial capital for exploring innovation activities [40]. Similarly, remittances enhance innovation through the channel of more foreign reserves and an increase in the income of the central government [41]. Both factors enable the government to advance the subsidies to the real sector of an economy for more innovation activities. The developed financial sector promotes innovation by offering the needed financing for innovation investment. At a macro level, the developed financial sector can accelerate innovation by making the import of modern technology more transparent [42]. However, high interest rates impede innovation by making the external financing of enterprises costlier, which discourages the enterprises from getting involved in innovation activities. Similarly, the high-interest rate enhances the burden of interest on existing debt advanced by the banking sector to the central government [43].

Lastly, subsidies stimulate innovation because they enable enterprises to allocate more funds for R&D activities and direct support from the government on innovation [51]. Subsidies play a crucial role in fostering green innovation by reducing the financial burden associated with research and development in environmentally friendly technologies. Government incentives, such as tax credits, grants, and direct funding, encourage firms to invest in cleaner production processes, renewable energy, and sustainable products. By lowering the cost of innovation, subsidies help mitigate the negative impact of air pollution on research productivity, enabling firms to adopt greener solutions. Additionally, well-designed subsidy programs can stimulate private sector participation, enhance technological spillovers, and drive long-term sustainability in innovation-driven economies.

While the study primarily examines the unidirectional impact of air pollution on innovation, it is important to acknowledge potential feedback loops that could further exacerbate environmental degradation. Reduced innovation, particularly in clean technologies and sustainable industrial practices, can slow down the development of pollution-mitigating solutions, leading to prolonged environmental harm. A decline in R&D investment in green energy, emission reduction technologies, and eco-friendly production processes could result in higher dependence on fossil fuels and inefficient industrial practices, creating a vicious cycle of worsening pollution and stifled innovation. This interplay highlights the need for proactive policies that simultaneously encourage technological progress and environmental sustainability to break this negative loop.

The findings of this study align closely with the broader discourse on sustainable development, particularly in the context of the United Nations SDGs. Air pollution’s detrimental impact on innovation performance directly relates to SDG 9 (industry, innovation, and infrastructure) by hindering technological progress and reducing R&D efficiency. Furthermore, the study underscores the environmental challenges tied to SDG 13 (Climate Action), emphasizing the need for stricter pollution control measures to foster a sustainable innovation ecosystem. By highlighting the intersection of economic growth, environmental sustainability, and innovation resilience, this research provides valuable insights for policymakers seeking to achieve sustainable industrial development while mitigating environmental risks.

Given that air pollution negatively impacts innovation, policymakers must implement stricter environmental regulations, promote clean energy adoption, and invest in pollution control technologies to sustain long-term economic growth. Additionally, targeted incentives for green innovation, such as tax breaks and subsidies for environmentally friendly R&D, can help mitigate the adverse effects of pollution on technological progress. For businesses, integrating sustainability into corporate strategies and adopting cleaner production processes can enhance competitiveness while aligning with global environmental standards. The study also offers a theoretical contribution by extending environmental economics frameworks to incorporate innovation performance, highlighting the need for a more integrated approach that links environmental quality, economic policies, and technological advancement.

While strict environmental regulations can play a crucial role in curbing pollution and fostering innovation, particularly in green technologies, they also present certain trade-offs. In the short run, stringent regulatory frameworks can increase compliance costs for firms, especially those in pollution-intensive industries. These costs may include investments in cleaner technologies, operational adjustments, and administrative burdens, which can reduce profit margins and potentially crowd out funds that would otherwise be allocated to R&D [52]. Thus, while environmental regulations are essential for sustainable development and long-term innovation, policymakers must carefully balance their design to avoid unintended negative impacts on firm competitiveness and innovation capacity in the short term.

6. Conclusions and Policies

This study highlights the critical challenge that air pollution poses to innovation performance in EAGLE economies, emphasizing the need to integrate environmental considerations into innovation and economic growth strategies. Drawing on robust econometric techniques (2SLS and FMOLS) and a 20-year dataset, the research provides compelling evidence that pollution—measured through CO₂, AP1, and AP2 emissions—significantly hampers R&D efforts, trademark activity, and patent applications by disrupting human capital, increasing financial uncertainty, and weakening the institutional environment necessary for innovation. Beyond its empirical contributions, the study advances theoretical understanding by situating the pollution-innovation relationship within the frameworks of endogenous growth theory, the Porter Hypothesis, and Institutional Theory. It shows how environmental degradation not only suppresses economic productivity but also erodes the core drivers of long-term innovation capacity.

To address these challenges, governments in EAGLE economies should strengthen environmental regulations by enforcing air quality standards and introducing incentive-based mechanisms such as pollution taxes or emission trading schemes. Promoting green financing instruments, including innovation bonds and low-interest R&D loans, can provide enterprises with the capital needed to pursue cleaner technologies. Building institutional capacity through transparent governance and efficient enforcement can create a stable environment that encourages innovation. Moreover, targeted investments in human capital, particularly in health and education, are essential to mitigate the productivity and cognitive losses caused by pollution. Finally, incorporating environmental quality indicators into national innovation policies will help align climate objectives with technological advancement, ensuring that innovation systems remain resilient and inclusive. Overall, coordinated policy actions are necessary to decouple innovation growth from environmental degradation and foster sustainable development in emerging economies.

6.1. Significance and Policy Implications

The findings highlight the critical need for context-specific policy interventions to mitigate air pollution and preserve innovation capacity in emerging economies. While environmental sustainability is essential for fostering innovation, generalized policy measures such as stricter regulations or green financing must be tailored to the economic and institutional contexts of individual countries. For instance, countries such as China can leverage their strong state-led innovation systems to invest in clean technology R&D through targeted subsidies and fiscal incentives, while economies such as Indonesia or the Philippines may benefit more from capacity-building initiatives that enhance local environmental governance and promote public-private partnerships for pollution control. In India, schemes such as the Perform, Achieve, and Trade (PAT) initiative, which incentivizes energy efficiency, offer a replicable example of aligning environmental goals with industrial performance. Similarly, Brazil could integrate environmental taxation mechanisms to disincentivize emissions while redirecting revenue toward innovation grants for green startups.

To ensure policy effectiveness, governments should adopt multi-pronged strategies that combine stringent emissions regulations with proactive support for innovation ecosystems. These could include tax credits for eco-innovation, air quality monitoring reforms, and improved transparency in pollution reporting. Furthermore, investing in public health systems to mitigate pollution-related health impacts is crucial to preserving the productivity of human capital. Cross-sectoral collaboration among ministries of environment, industry, health, and education, as well as partnerships with international agencies and regional networks, will also be vital for knowledge-sharing and mobilizing technical and financial resources. By aligning environmental and innovation policies more closely with national development goals, emerging economies can build resilient innovation systems that thrive even under environmental constraints.

The findings of this study offer valuable insights for informing international collaborations and agreements such as the Paris Climate Accord by highlighting the dual challenge of mitigating air pollution while sustaining innovation-led growth. The demonstrated negative impact of pollution on innovation outcomes suggests that environmental degradation not only undermines climate goals but also hampers countries’ technological progress and economic resilience. This underscores the need for integrating innovation support mechanisms into global climate frameworks. For instance, under the Paris Accord’s climate finance and technology transfer provisions, developed countries can support emerging economies in adopting cleaner production methods and scaling up eco-innovation. Moreover, the study reinforces the case for establishing international knowledge-sharing platforms and cooperative R&D programs focused on green technologies, enabling countries to collectively bridge the innovation gap. These collaborations can also facilitate harmonized pollution standards, shared monitoring systems, and joint investment in cross-border green infrastructure, ensuring that environmental sustainability and innovation are pursued in tandem on a global scale.

6.2. Limitations and Future Research Agenda

While this study sheds light on the negative impact of air pollution on innovation performance within EAGLE economies, several limitations warrant consideration. The study’s focus on R&D expenditures and patent applications as proxies for innovation might overlook other facets of innovation, such as qualitative aspects or non-patentable innovations. Furthermore, it does not conduct subgroup analyses to account for potential heterogeneity stemming from differences in economic structures, regulatory frameworks, or pollution intensities among countries such as China and Brazil. This lack of disaggregated analysis represents a limitation, as the uniform treatment may overlook country-specific dynamics that could influence the strength or direction of the observed relationship. Future research could address this by exploring heterogeneity across income levels, industrial compositions, or environmental governance standards. Additionally, longitudinal studies assessing the long-term adaptation strategies of firms facing environmental constraints would provide deeper insights. Investigating the role of green policies and clean technology adoption in mitigating pollution’s adverse effects on innovation could also be a valuable avenue for further study. Additionally, employing more nuanced methodologies that capture the multifaceted nature of innovation and environmental impacts would provide a more comprehensive understanding of how air pollution affects innovative activities within emerging economies. Moreover, investigating the role of specific environmental policies (environmental taxation, etc.) or interventions in mitigating the adverse effects of air pollution on innovation could offer valuable insights for policymakers seeking effective strategies to promote innovation amidst environmental challenges.