< PreviousOwen Cheung and Semee Yoon •324• the same time, nearly two thirds of India’s cities had annual levels of PM 2.5 that were more than seven times greater than the levels recommended by the World Health Organization’s guidelines (IQAir, 2022; WHO, 2021). In 2019, air pollution led to the death of 1.67 million Indians, which represented a 17.8% share of total mortality for the year (Pandey et al., 2021). A loss of life of such a magnitude constitutes a public health crisis. That loss was not only felt in strictly human terms, but in economic terms as well. The lost output attributable to premature deaths in that year was estimated at $28.8 billion, with morbidity accounting for an additional $8.0 billion (Pandey et al., 2021). These figures serve as reminders that loss of life, no matter how it is measured, is a persistent tragedy. Given the gravity of the situation, there has been growing research on the impact of evolving policy efforts in India. The issue of air pollution is not a new one in India. Since 1981, various government efforts have defined the country’s air quality preservation landscape. The Air Prevention and Control of Pollution Act was the first of such efforts to set air quality standards (Ministry of Environment, Forest & Climate Change Govern- ment of India, 2019). This was followed by the National Ambient Air Quality Standards (NAAQS) in 1982. NAAQS compliance is determined by both daily and annual levels of air pollutants. The acceptable standards for PM less than 10 and 2.5 micrometers are set through NAAQS and have been revised throughout the years in accordance with updated scientific research and air pollution reduction goals (Somvanshi and Roychowdhury, 2020). The idea of abiding by certain standards for air pollution is integral to India’s approach to air pollution. This long history of setting air quality standards suggests that the Indian government recognizes why air pollution is a serious problem. The fact that poor quality persists suggests that the issue was that Indian cities have not been able to stay within these standards; indeed, this is the rationale for the creation of NCAP, as will be discussed further in the paper. Recent literature on the air pollution in India has not delved into components of NCAP. Thus, we contribute to the literature by exploring the effect of the underutilization of allocated funds by NCAP’s “non-attainment” target cities on air pollution reduction outcomes.The Effect of the Underutilization of Funds by Non-Attainment Cities in India’s National Clean Air Programme •325• The next chapter provides the literature on policy-led pollution initiatives and recent findings on policy changes in India. Chapter 3 provides the research context in India. Chapter 4 describes the methods and Chapter 5 presents results and discussion. The last chapter concludes with future research directions. Ⅱ. Literature review 1. Previous findings on the effectiveness of policy-led pollution control initiatives The Clean Air Act (CAA) is the most notable policy effort in the United States. Originally enacted in 1970 with amendments in 1977, 1990, and afterwards, CAA is a federal program to strengthen the federal government’s role in pollution reduction (Currie and Walker, 2019). The CAA established the Environmental Protection Agency (EPA) charged with mitigating air pollutants. The EPA requires states to submit annual plans to make “nonattainment” counties compliant with air quality standards and can withhold related funding to states that do not submit plans. Since its introduction, the CAA has led to significant reductions in air pollution across the US (Samet, 2011), and studies have shown the policy to have been highly cost-effective (Krupnick and Morgenstern, 2002). The CAA represents a historical example of how a federal program can engage local-level governments in pollution reduction. In the European Union, the Convention on Long-range Transboundary Air Pollution (LRTAP) was signed in 1979, 9 years after the CAA. Unlike the American approach, the EU implemented supranational regulations on the emissions of air pollutants, which led to a reduction in SO 2 and NO x emissions (Kuklinska et al., 2015). Further directives passed targeted other chemical compounds with varying degrees of success. In comparison to the US policy, the EU’s policies are a mixture of EU-wide directives and member-states’ own Owen Cheung and Semee Yoon •326• standards and strategies which may be even more stringent as in the cases of Austria and the United Kingdom (Kuklinska et al., 2015). In more recent years, China has been targeting reductions in PM 2.5 levels in areas with severe air pollution through its Air Pollution Prevention and Control Action Plan. Since 2013, the Action Plan mandated significant reductions in annual emissions and has thus far been effective (Zhang et al., 2019). The policy successfully reduced PM levels in targeted regions between 2013 and 2017. Additionally, the reductions led to significantly fewer deaths and years of life lost (Huang et al., 2018). However, the effects of the policy may be limited to the regions in which the reductions are mandated, and may also be dependent on good management of bureaucratic systems (Zeng et al., 2019). Furthermore, China’s progress in PM 2.5 and PM 10 reductions has coincided with worsening ozone levels (Mousavinezhad et al., 2021). 2. Research findings on air pollution in India Urbanization and industrialization have led to the air pollution problem faced by India today. Densely concentrated human activities in urban areas have been shown to be associated with increased air pollution in the form of ambient PM 10 , PM 2.5 , NO 2 , and SO 2 levels (Liu et al., 2022). In Delhi, one of India’s most populous cities, modeled data finds that vehicle exhaust, the burning of biomass, and soil and road dust contribute the most to PM 2.5 , while industry, the burning of waste, diesel generators, and power plants also contribute to ambient PM 2.5 pollution (Guttikunda et al., 2023). These causes constitute significant but identifiable challenges that the Indian government eventually sought to address in NCAP.The Effect of the Underutilization of Funds by Non-Attainment Cities in India’s National Clean Air Programme •327• Ⅲ. National Clean Air Programme in India Recognizing the direness of the air quality situation, the Indian government passed the National Clean Air Programme (NCAP) in 2019, with the intention to “ensure [air pollution’s] decoupling with the rapidly growing economy” (MoEFCC, 2019). In the introduction of the NCAP in 2019, India’s Ministry of Environment, Forest and Climate Change (MoEFCC) emphasized the importance of rapid economic growth to the Sustainable Development Goals, including SDG 1 No Poverty, SDG 2 Zero Hunger, and SDG 3 Good Health and Well-being, among others (MoEFCC, 2019; UN, n.d.). However, they acknowledged that such economic growth also leads to increased emissions, leading to negative externalities associated with poor air quality. The Ministry also acknowledged that many cities exceeded air quality standards due to local, regional, and trans-boundary sources, which motivated the development of the program (MoEFCC, 2019). The NCAP is the first effort on the national level in India to try to achieve a time-bound reduction target for air quality management; by 2024, cities were required to reduce their ambient PM concentration by 20-30 percent (Somvanshi and Roychowdhury, 2020). The program operates at both the national level in which various policies are implemented and enforced by the centralized government, and somewhat uniquely, at the municipal level (MoEFCC, 2019). This tiered approach is designed to empower cities to address their unique sources of air pollution, which vary from area to area depending on the industrial make-up, energy sources, road infrastructure, and other factors (Somvanshi and Roy- chowdhury, 2020). Variations in the concentrations of air pollutants are also affected by meteorological parameters in Indian cities, e.g. the strength and direction of the wind (Suthar et al., 2023). Such variations make each locality’s air pollution situation dependent not only on sources of pollution, but how that pollution is dispersed and concentrated. As such, the Central Pollution Control Board (CPCB) of the central government designated certain cities as so-called “non-attainment cities.” The first 94 cities were Owen Cheung and Semee Yoon •328• designated based on their exceeding of NAAQS standards for PM 10 before 2015. Then, 16 cities were added after 2015 as they also exceeded separate PM 2.5 standards. The World Health Organization also identified 10 more cities based on its own data, and in the process of implementing NCAP, 20 more cities were included based on updated data collected by Indian government (Somvanshi and Roychowdhury, 2020). By November 2023, there were 131 cities that have not met the NAAQS air quality standards for PM and that receive funding under NCAP (CREA, 2024). The non-attainment cities are obligated to develop and enact Clean Air Plans that address the sources of air pollution at the local level. The integration of action plans designed by local stakeholders is also found in the strategies of developed countries; local action plans have been used in England since the beginning of the millennium (Beattie et al., 2002; Beattie et al., 2006). However, the evidence of the air pollution reduction effect of the plans is unclear (Barnes et al., 2015). More recently, in California’s Assembly Bill 617, Community Emissions Reduction Plans are to be submitted to an oversight committee (Fowlie et al., 2020). These plans, like NCAP’s, are meant to develop specific emissions reduction strategies through the identification of localized sources of emissions and pollution. Thus, a well-intentioned, concerted effort to reduce pollution levels began at both the national and municipal levels. However, myriad challenges have beguiled the program since its inception, leading to extensions of reduction targets and delays in the establish- ment of air pollution forecasting systems (CREA, 2024). At the initial stages of NCAP, key challenges that were raised were regarding legal foundations to implement the plan by cities. The Council on Energy, Environment and Water (CEEW), a public policy think tank in India, pointed out that there is a lack of a legal mandate to review and continually update the clean air plans (Ganguly et al., 2020). This concern seems to have been addressed by the central government; by 2023 the government was requiring the submi- ssion of an annual action plan by non-attainment cities subject to central government approval (Central Pollution Control Board, 2023). The Effect of the Underutilization of Funds by Non-Attainment Cities in India’s National Clean Air Programme •329• The more recent challenge that has emerged is the widespread underutilization of allocated funds, which despite being released to cities, were not actually used (CREA, 2024). Through the program, the central government allocates funding for the Clean Air Plans to the cities themselves, which means that the burden of funding does not fall upon municipal governments (MoEFCC, 2019). Indeed, most of the costs of the program itself fall upon the central government rather than the state or city governments. Thus, the onus placed upon the cities is not financial, but for implementation. Given that both the causes and negative impacts of air pollution are localized to and more severe in cities, pollution reduction efforts naturally need to be implemented at the municipal level. An annual report by the Centre for Research on Energy and Clean Air found that until November 2023, among the 82 originally designated non-attainment cities, 498 crores 1) of the allocated 1253 crores of rupees were utilized by the cities, which represents a mere 40% utilization rate. Among the other 49 cities, 9610 crores were allocated and 5909 crores were utilized by those cities, a higher but still disappointing 62% utilization rate (CREA, 2024). Rather than budgetary constraints that limit cities from enacting their air pollution reduction agenda, the problem is a lack of participatory will. Still, there are signs that NCAP is working to improve urban air quality in India. One study found that surface ozone pollution has been reduced since the program has started (Gopikrishnan and Kuttippurath, 2024). Although surface ozone is not the primary target of NCAP’s reduction goals, it is a positive sign that NCAP has been making an impact. In Chandigarh, one of the non-attainment cities, the concentrations of PM and CO increased from 2021 to 2022, but the rate of increase was limited by the measures enacted by the city under NCAP (Dhote et al., 2024). Despite the absence of a reduction in air pollution, the study demonstrates that NCAP still may have a positive role for India’s air quality even if reduction targets are not met by 2026. Another study examined NCAP’s implementation in Ahmedabad, finding a 7.2% reduction in PM 2.5 levels from 2018 to 1) A crore is equivalent to 10 million rupees, around 120,000 USD or 160,000,000 KRW as of 2025.Owen Cheung and Semee Yoon •330• 2022 (Kapoor et al., 2024). They estimated that the reduction in PM exposure led to 1631 fewer deaths in the city during that period (Kapoor et al., 2024). These findings demonstrate that NCAP has the potential to create meaningful reductions in air pollution in India’s most polluted cities. To further evaluate the impact of NCAP, we aim to understand the effect of under- utilization of funds by city governments on air quality. Existing literature has demonstrated that spending by the government leads to improvements in air quality. Halkos and Paizanos (2017) found that government expenditure leads to an alleviation of SO 2 and NOx levels, whereas Bernauer and Koubi (2013) found that larger government expenditure was correlated negative relationship with air quality, meaning that larger governments had worse air pollution. However, when government spending is specifically earmarked for the environment, the relationship becomes positive. He et al. (2018) linked environ- mental expenditure to improvements in the air quality index among 7 cities from 2007 to 2015. Additionally, Huang (2021) found a similar relationship between environment government spending and improvements in air quality in Taiwan. The paper also found that subsidies by the central government to local government had the greatest effect among funding sources, which is the basis of NCAP’s funding structure. Prior research emphasizes the importance of targeted spending on air pollution for achieving improvements in air quality. Thus, underutilization of allocated funds in NCAP may be a critical aspect to analyze to present the effectiveness of such air quality initiative. Ⅳ. Materials and Methods In a country where life expectancy is reduced by 5.3 years because of particulate pollution, the importance of the success of NCAP cannot be understated (Air Quality Life Index, n.d.). The underutilization of funds by non-attainment cities creates a natural experiment. Using data on the utilization level of allocated funds and localized air quality, The Effect of the Underutilization of Funds by Non-Attainment Cities in India’s National Clean Air Programme •331• this research seeks to determine the effect of the implementation of non-attainment cities’ plans to improve air quality. If the NCAP’s strategy of empowering cities to establish their own plans to achieve significant reductions in particulate pollution is viable, then cities that have higher levels of utilization should have made gains in pollution reduction compared to cities with low or no utilization of allocated funds. Data for this research was sourced through the CPCB’s Portal for Regulation of Air Pollution in Non-Attainment Cities (PRANA) (Central Pollution Control Board, 2024). However, the CPCB does not provide compiled datasets so the data for 131 cities was collected manually from each city’s individual portal page. Figure 1 shows how funding, utilization, and air quality data is provided to the public by the government on the PRANA website. Each of these pages includes data for the funds allocated by the central government and funds utilized by the city from 2019 to 2023. Additionally, each page includes annual averages of PM 10 levels from 2017 to 2023. Some cities also received funding via the Central Government’s Fifteenth Finance Commission (XV-FC) grants to Urban Local Bodies (Central Pollution Control Board, 2022). The funds provided by XV-FC grants are also reported on the PRANA website. Similar to NCAP, these grants are also intended to be used to reduce PM levels in cities (MoEFCC & Government of India, 2023). As dual NCAP and XV-FC cities (n = 49) have separate funding coming from both sources used for the same purpose, it is difficult to estimate the isolated effect of NCAP’s funding on air quality. Consequently, they have been removed from the study, which represents a potential source of selection bias (MoEFCC & Government of India, 2023). The XV-FC grants are largely provided to cities with populations above one million. Thus, the number of cities included in this study is 82, which are primarily cities with populations below 1 million people.Owen Cheung and Semee Yoon •332• <Figure 1> Funding and PM 10 Data in Non-Attainment Cities Source: Central Pollution Control Board (2024). Given that the funding and air quality data are available annually, we ran a panel data regression to explore the effect of funding levels on air quality improvements. We use a fixed effects (FE) model with robust standard errors. As air quality is affected by city- specific time-invariant characteristics, data which are largely unavailable in the Indian context, the FE approach is advantageous because it controls for these city-level factors. In contrast, a pooled OLS model assumes that those unobserved city characteristics do not influence pollution or fund utilization rate, which is unlikely due to the heterogeneous nature of Indian cities. Statistical tests were performed to support the use of a FE model. We first compared random effects (RE) with FE using a Hausman test. This resulted in a chi-square test statistic of 19.418 and a p-value of 0.002 (Appendix A), which confirms that city-level characteristics are correlated with the explanatory variables. This result indicates that the RE model would yield inconsistent estimates, while the FE model provides unbiased estimates. Therefore, we use FE as our primary specification.The Effect of the Underutilization of Funds by Non-Attainment Cities in India’s National Clean Air Programme •333• To test for heteroskedasticity in our panel dataset, we used the Modified Wald test for groupwise heteroskedasticity as presented in Appendix B. The test rejected the null hypothesis (χ² = 4.1e + 05, p = 0.000), indicating that variance across cities is not constant. This confirmed the presence of heteroskedasticity. To address this, we use heteroskedasticity-robust standard errors in our FE model. Additionally, to check for serial correlation, we conducted the Wooldridge test for autocorrelation in panel data, shown in Appendix C. The test fails to reject the null hypothesis (F(1, 81) = 2.115, p = 0.1498), indicating no strong evidence of first-order serial correlation in the dataset. This suggests that errors are not systematically correlated across time within cities. As a result, clustered standard errors are not necessary, though we continue to report heteroskedasticity-robust standard errors to ensure the validity of our estimates which include city clusters. The equation for the estimated regression is the following: ∆ ∆ ∆ ∆ (1) The final model uses the change in PM concentrations year-to-year ( ∆ ) as the outcome variable because rather than absolute pollution levels, we are interested in how funding affects changes in air quality. The average change in PM was a decrease by 2.994 micrograms per cubic meter (Table 1). The main independent variable of interest is the use rate of funds by the city. This is calculated as the funds utilized by the city during the fiscal year in proportion to the funds released to the city by the central government. Cities use funds received for pollution source appropriation studies, the expansion of air quality monitoring capacity, and various air pollution reduction efforts in accordance with their individual city action plans. Additionally, the CPCB stipulates that cities spend no more Next >