The insufficiency of national climate policies to avert severe climate risks has led to the emergence of transnational municipal networks (TMNs) as potential actors in climate change mitigation. TMNs, comprising city governments, aim to bypass national efforts by directly providing expertise and resources to member cities to reduce greenhouse gas emissions. While some argue TMNs offer a bottom-up approach fostering innovation and collaboration, particularly beneficial for developing countries lacking technical expertise, others express skepticism, suggesting TMNs may prioritize established solutions and disseminate information selectively to advance political agendas. This study focuses on a critical emissions source: power plants. Existing research lacks global data on individual power plant CO₂ emissions, preventing a comprehensive assessment of TMNs' impact. The study aims to fill this gap by analyzing the association between TMN membership and power plants' CO₂ emissions, considering national policies and developmental context, and focusing on whether TMNs effectively reduce emissions from the highest-polluting plants.

Existing research on TMNs has examined various aspects, including member city experiences, comparisons of well-known TMNs, factors driving city participation, and the ambition levels of municipal climate actions. Several studies have investigated the policy instruments used by TMNs and the sectors they target. However, research directly assessing TMNs' impact on CO₂ emission reduction has been limited. Leffel's work demonstrated a link between TMN memberships and greater city-level GHG reductions, even when controlling for national climate policy, suggesting TMNs facilitate bottom-up diffusion of climate policy knowledge. This contrasts with world society theory which traditionally emphasizes a top-down approach. Conversely, concerns have been raised regarding the effectiveness of TMNs, with criticism focusing on their potential to promote politically expedient solutions, selectively disseminate information, and replicate existing solutions from developed countries, potentially overlooking the specific needs of developing nations. The limited existing research on TMN’s effect on pollution has primarily focused on downstream consumers rather than upstream producers such as the electricity sector. This study addresses this gap by specifically examining the electricity sector, highlighting its pivotal role in climate-smart cities and the importance of decarbonizing the most egregious polluting power plants.

This study uses a novel international dataset comprising information on individual power plants, their CO₂ emissions (2009-2018), technical specifications, national energy-climate policies, and TMN membership status of the cities housing the plants. The dependent variable is the power plant's CO₂ emission level in 2018, log-transformed to address skewness, with a lagged 2009 emission level as a control. Data on plant emissions is drawn from an updated Carbon Monitoring for Action (CARMA) dataset, combining various sources including plant-level emission reports and statistical estimations for non-reporting plants. The key independent variable is a dummy variable indicating TMN membership (with a focus on the electricity sector), sourced from the Carbon Disclosure Project. Alternative operationalizations of TMN involvement, using total years of experience and longest membership duration, were also tested but proved insignificant. Controls include factors such as fuel source (coal), plant capacity, capacity factor, plant age, government utility status, GDP per capita, inflation, unemployment, population change, national fossil fuel power capacity, climate risk, national energy consumption, number of NGOs, degree of liberal democracy, and an indicator of national energy-climate policies from the International Energy Agency. Given the nested structure of the data (plants within countries, sub-national areas, and parent companies), a hierarchical linear mixed-effects model with three random intercepts (for countries, sub-national areas, and parent companies) was employed to account for this nesting and the unbalanced design.

The analysis revealed a significant negative association between TMN membership and power plants' CO₂ emissions. Plants in cities participating in TMNs emitted less CO₂ over time, even after controlling for structural properties and national energy-climate policies. This effect was particularly pronounced in less developed countries, indicating TMNs potentially provide access to decarbonization resources otherwise unavailable. However, TMNs showed no significant relationship with emission reductions among the highest-polluting plants (top 10% and 25%). National energy-climate policies had a negligible impact on plants' environmental performance. The results are summarized in Table 1, which presents the results of four hierarchical linear mixed effects models. Model 1 shows the overall effect of TMNs on CO2 emissions. Model 2 investigates the interaction effect of TMNs and less developed countries. Models 3 and 4 examine the interaction effects of TMNs with the top 10% and top 25% of polluting plants, respectively. The table shows the coefficients, standard errors, and significance levels for each variable in the models. Figure 2 graphically presents the interactions examined in Models 2-4, showing point estimates and 95% confidence intervals.

The findings support arguments both for and against TMN effectiveness. The positive impact of TMNs on reducing CO₂ emissions, particularly in developing countries, suggests TMNs provide crucial climate policy support and access to decarbonization resources. This is further supported by qualitative examples of cities such as Jakarta, Manila, Hanoi, and Sao Paulo, demonstrating how TMN membership facilitated or accelerated energy sector decarbonization efforts. The lack of impact on the highest-polluting plants, however, suggests TMNs may not be effective in addressing the most significant sources of emissions. The greater impact of city-level TMN factors compared to national policies underscores the importance of polycentric governance structures, particularly in bridging resource gaps in developing economies. The study highlights the need to investigate TMNs' effects on power plant decarbonization beyond aggregate city-level outcomes.

This study provides the first global empirical evidence that TMNs can reduce the electricity sector's CO₂ emissions, surpassing the influence of national policies. The significant effect of TMNs in less developed countries underscores their role in filling critical policy support gaps. However, their limited effect on the most egregious polluters necessitates further research to understand the factors mediating TMN effectiveness. Future studies should investigate the role of city capacity, resource availability, and the tailoring of policy analysis offered by different TMNs.

The study acknowledges the limitation that only a small percentage of power plants are located in cities belonging to TMNs, which could limit the generalizability of the findings. The reliance on existing datasets might also introduce limitations concerning data availability and quality. The definition of TMN involvement is important and the researchers primarily focused on the power sector to limit scope and increase precision in their evaluation.