Climate Conferences and the Ballot Box

As COP31 opens in Türkiye, the American midterm elections raise a difficult question. Are international climate negotiations or public resistance more effective in changing government policy?

On November 3, 2026, American voters went to the polls for the midterm elections. In dozens of American communities, residents had organized petitions to place restrictions on data-center development on the November ballot. The elections provided an opportunity for communities to express their opposition through direct democratic procedures.

Opposition to these facilities had been growing for months, driven by concerns over electricity prices, water consumption, environmental damage, and the expanding influence of technology companies over local economic decisions.

Six days later, on November 9, the thirty-first United Nations Climate Change Conference (COP31) is scheduled to open in Antalya, Türkiye.

The previous thirty conferences have produced agreements, targets, financing mechanisms, and institutional arrangements. Yet, global greenhouse gas emissions remain far above the levels consistent with the Paris Agreement. Meanwhile, opposition to data centers in the United States has moved from public meetings and demonstrations to local referendums and electoral campaigns.

Does political change depend more on international negotiations or on citizens making their dissatisfaction felt at the ballot box?

International agreements can establish common objectives. However, their implementation depends on domestic political and economic conditions. Public resistance can alter those conditions, although it does not necessarily produce better environmental policies.

Thirty Climate Conferences

The United Nations Framework Convention on Climate Change was adopted in 1992, and the first Conference of the Parties was held in Berlin in 1995. The purpose was to create a framework through which governments could coordinate their responses to a problem that no country could solve independently.

The subsequent negotiations produced several major achievements.

The Kyoto Protocol, adopted in 1997, established binding emission-reduction targets for participating industrialized countries. Its coverage was limited. Major emitters either did not participate fully or were not subject to comparable obligations.

The Paris Agreement, adopted at COP21 in 2015, brought countries into a common framework for limiting global warming to well below 2°C above pre-industrial levels while pursuing efforts to limit it to 1.5°C. Governments agreed to submit nationally determined contributions (NDCs), strengthen them over time, and report on their progress.

Subsequent conferences expanded the framework. COP27 in Egypt established arrangements for a loss-and-damage fund to assist vulnerable countries. COP28 in Dubai produced the first global stocktake under the Paris Agreement and called for a transition away from fossil fuels in energy systems. It also called for tripling global renewable energy capacity and doubling the annual rate of energy-efficiency improvement by 2030.

At COP29 in Azerbaijan, governments agreed on a new climate-finance goal of at least $300 billion annually by 2035 for developing countries, with a broader ambition to mobilize $1.3 trillion annually. COP30 in Brazil advanced negotiations on adaptation finance and a just transition.

These developments, documented by the UNFCCC, have created international standards, reporting obligations, and mechanisms for cooperation. Nevertheless, an agreement is only the first step toward solving a problem.

The Evidence of the Emissions Gap

The United Nations Environment Programme's Emissions Gap Report 2026: The Point of Return, published on October 21, assesses how far governments remain from their stated objectives.

Under existing policies, global warming could reach approximately 2.8°C during this century and continue increasing thereafter. Full implementation of current national climate commitments, together with comparable efforts thereafter, would limit warming to approximately 2.6°C. Even the more optimistic scenario in which governments fulfill their NDCs and additional net-zero commitments produces an estimated warming of 2.1°C.

These projections are conditional estimates based on different assumptions about policy implementation, emissions, and long-term climate responses. Their significance lies in the distance between existing commitments and the emissions pathways required to meet internationally agreed temperature goals.

UNEP calculates that a pathway limiting the eventual overshoot of 1.5°C would require global greenhouse gas emissions to fall by approximately 18% by 2030, 45% by 2035, and 69% by 2040, relative to 2025 levels. Even this pathway would involve a temporary temperature increase to approximately 1.7–1.8°C around the middle of the century, followed by a return to 1.5°C or below by 2100 with an estimated probability of around 50%.

By comparison, full implementation of existing NDCs would reduce emissions by only 5–8% by 2030 and 11–15% by 2035, relative to 2025.

Governments have agreed on the objective, but their collective commitments remain inconsistent with achieving it.

There has nevertheless been progress. UNEP acknowledges that renewable energy deployment has expanded considerably, climate policies have produced measurable emissions reductions in several countries and sectors, and projected warming is lower than estimates made around the time of the Paris Agreement.

The difficulty is establishing how much of this progress resulted directly from international negotiations rather than technological improvements, declining renewable-energy costs, national legislation, or changing investment patterns.

The COP process has contributed to climate governance without securing emissions reductions on the scale required by its own objectives.

The Paris Agreement relies heavily on nationally determined commitments. Although it establishes binding procedural obligations, it does not impose internationally enforceable emission-reduction quotas on individual countries. Governments retain considerable discretion over the design and implementation of their policies.

Consequently, climate commitments are filtered through domestic politics, industrial interests, fiscal constraints, employment considerations, and international competition.

This brings us to the American midterm elections.

Data Centers

The rapid expansion of artificial intelligence has increased demand for computing infrastructure. Data centers have become central to the investment strategies of major technology companies, including Microsoft, Google, Amazon, Meta, and Oracle.

Their expansion has created new demands on electricity networks, water resources, and land. According to the International Energy Agency's Energy and AI report, published in 2025, data centers consumed approximately 415 terawatt-hours of electricity worldwide in 2024, equivalent to 1.5% of global electricity consumption.

The IEA projects that consumption could reach approximately 945 terawatt-hours by 2030 in its base scenario, more than doubling within six years. The United States accounts for the largest share of existing data-center electricity consumption and is expected to experience the largest absolute increase.

The global percentage may appear relatively modest, but the geographical concentration of demand matters. A large data center can consume electricity on a scale comparable to an energy-intensive industrial facility. Its location can affect the capacity requirements of regional electricity networks, the construction of new generating facilities, and the allocation of infrastructure costs.

The environmental consequences depend on how the additional electricity is generated, the technologies used for cooling, local water availability, and the characteristics of the electricity grid. Data centers are not environmentally identical, and their construction does not necessarily lead to equivalent increases in fossil-fuel consumption.

Technology companies receive the commercial benefits of expanding computing capacity, while surrounding communities may bear some of the costs associated with electricity infrastructure, water consumption, land use, and environmental disruption.

The question is who determines their location, who benefits from them, and who pays for the infrastructure and environmental consequences.

In an August 26 report, Associated Press documented how opposition to data centers had brought together constituencies that would ordinarily have little political common ground. Conservative farming communities, environmental organizations, and residents concerned about rising utility costs found themselves opposing similar projects for different reasons.

The conflict also exposed divisions within organized labor. Construction unions often supported data-center investment because of the employment opportunities it created, while environmental organizations and local residents questioned the projects' long-term consequences.

Opposition to data centers is not a unified environmental movement. It includes climate concerns, economic grievances, objections to land-use decisions, and resistance to the concentration of corporate power. These grievances share a concern that the costs of economic transformation are being imposed on communities without sufficient democratic participation in the decisions.

From Public Resistance to Electoral Results

Before the November elections, opposition to data centers had already become a significant political issue.

According to Reuters' October 29 reporting, voters in 38 local jurisdictions across ten states were preparing to decide at least 41 measures intended to delay, prohibit, or impose additional requirements on data-center development. Ohio was particularly prominent, accounting for 21 data-center-related proposals.

The results demonstrated that the opposition had acquired considerable electoral strength.

A Reuters analysis published on November 6 reported that 40 of the 41 ballot measures tracked by Redbud Consulting favored restrictions on data-center development. The results were spread across ten states, indicating that the issue was not confined to one political constituency or geographical region.

The November 7 analysis by Al Jazeera similarly found that voters had rejected unrestricted data-center development in virtually all the contests it examined. The exception was Harwood, North Dakota, where voters rejected a proposal requiring independent reviews of large industrial projects.

Some measures sought outright prohibitions, while others proposed moratoriums, additional reviews, or stronger regulatory requirements. A vote for restrictions cannot automatically be interpreted as opposition to all data-center development.

Communities that had expressed concerns through protests and local political campaigns were now using formal democratic procedures to influence development decisions.

The results also had implications beyond local referendums.

Reuters reported that counties holding data-center ballot measures frequently experienced shifts toward Democratic candidates, including in important Senate contests in Ohio and Kansas. Several successful candidates had incorporated data-center regulation into their campaigns, while the prospect of a more interventionist Congress increased expectations of tighter oversight of the technology industry.

The fact that voters supported restrictions on data centers and also voted for Democratic candidates does not establish that opposition to data centers caused the broader electoral shift. Voters were simultaneously responding to inflation, national economic policy, employment concerns, and numerous other political issues.

The November results establish that resistance to data centers became electorally consequential, particularly at the local level. They do not yet establish the number of congressional or gubernatorial races that would have produced different outcomes in the absence of that resistance.

Do Protests Actually Change Political Outcomes?

The relationship between protests and political change has been examined in a growing academic literature. Its findings are more complex than the assumption that public demonstrations necessarily lead to policy reform.

In a 2025 review published in Current Opinion in Behavioral Sciences, researchers examined 50 studies investigating the effects of climate activism.

Their review distinguishes between changes in public opinion, media attention, voting behavior, political communication, corporate behavior, environmental policy, and emissions.

The evidence is strongest for the effects of activism on public opinion and media coverage. The authors also find moderate evidence that climate activism can influence voting behavior, encourage politicians to address climate issues, and exert financial pressure on carbon-intensive companies.

The evidence becomes less conclusive when the analysis moves from political attention to actual environmental outcomes. The authors find suggestive evidence that activism can influence policy and emissions, but emphasize that much of the existing research concentrates on intermediate outcomes rather than long-term environmental changes.

A protest may attract attention without changing voting behavior. It may change voting behavior without producing legislation. Legislation, in turn, may fail to deliver the intended reduction in emissions.

The authors also identify a geographical limitation. Much of the research concerns the United States and Western Europe, making it difficult to generalize its findings to other political and institutional settings.

Protests can also produce political reactions contrary to those intended by their organizers.

A 2026 study published in the European Journal of Political Economy, titled Backlash at the Ballot Box: Electoral Repercussions of Climate Activism, examined the Stop Adani campaign in Australia.

Before the 2019 Australian federal election, environmental activists organized a convoy opposing the development of a major coal mine. Using a quasi-experimental research design and difference-in-differences analysis, the researchers investigated how exposure to the convoy affected electoral outcomes.

They found that exposure increased support for the conservative Coalition and reduced support for the environmentalist Greens. The effects were concentrated in areas economically dependent on mining, particularly in Queensland.

The findings suggest that the political consequences of environmental activism depend partly on how voters perceive its economic costs. Where communities believe that environmental restrictions threaten employment and livelihoods, protests may strengthen opposition to environmental policies.

This result is relevant to the American data-center controversy. In some communities, data centers are perceived as a threat to affordable electricity, water resources, and local control. Elsewhere, they are regarded as sources of construction employment, investment, and tax revenue.

The same technology can generate opposing political coalitions. Whether protests produce electoral support or backlash depends on the distribution of expected benefits and costs.

Why International Commitments Encounter Domestic Resistance

Climate change is a global problem, but the authority to regulate economic activity remains largely national and local.

Governments negotiate international commitments while simultaneously pursuing economic growth, industrial competitiveness, energy security, and employment. Conflicts arise when the costs of transition are concentrated in particular industries, regions, or social groups.

A government may support ambitious international climate targets while permitting investments that increase domestic energy demand or prolong dependence on fossil fuels. It may promote renewable energy while continuing to subsidize carbon-intensive activities. It may announce long-term emissions objectives without adopting sufficiently strong short-term measures.

These contradictions reflect conflicts among economic interests, institutional constraints, and political incentives. The Paris Agreement expects governments to strengthen their commitments over time, but the international system has limited capacity to compel them to implement particular emission-reduction measures.

The gap between commitments and implementation is a political problem.

Residents confronting a proposed data center may experience its anticipated consequences through electricity bills, water availability, land use, noise, or the prospect of new power plants. These concerns are immediate and geographically concentrated.

The consequences of climate change are distributed across countries, generations, and ecosystems. Although its effects are increasingly visible, the relationship between individual policy decisions and global temperature outcomes is less direct.

A local electorate can reject a development proposal or remove an elected official. It cannot compel other countries to reduce emissions.

International negotiations remain necessary for coordinating global action, while domestic political mobilization can influence whether governments implement their commitments.

Can Electoral Resistance Produce Better Climate Policy?

The American midterms demonstrate that opposition to data centers can influence development decisions. Whether such opposition will improve climate outcomes is less certain.

A rejected data-center project may be relocated to another jurisdiction rather than abandoned. Its electricity demand may remain unchanged. If the alternative location relies on more carbon-intensive electricity generation, the environmental consequences could even be worse.

Conversely, restrictions requiring developers to finance additional electricity infrastructure, improve water efficiency, or secure low-carbon energy supplies may reduce environmental costs without preventing investment altogether.

The relevant measure of success is not the number of projects rejected but the environmental and economic consequences of the policies adopted.

Public resistance can influence the political agenda by forcing elected officials to address neglected concerns. Electoral pressure may then produce changes in legislation, planning requirements, utility regulation, and corporate obligations. Whether these changes reduce greenhouse gas emissions, water consumption, and environmental damage, or improve electricity affordability, requires separate assessment.

The first two developments are increasingly visible in the American data-center controversy. The third will require longer-term evaluation.

The same standard applies to international climate conferences. Their success must ultimately be assessed through the implementation of policies and their effects on emissions, climate risks, and the ability of societies to adapt.

What Should COP31 Learn from the American Elections?

COP31 begins at a time when international climate governance faces growing pressure to demonstrate results.

UNEP's 2026 assessment shows that existing national commitments remain insufficient to meet the Paris Agreement's temperature objectives, despite substantial progress in renewable energy and climate-policy development.

The American elections demonstrate that economic and environmental concerns can become politically decisive when citizens have opportunities to influence specific decisions.

Academic research confirms that activism can affect political behavior, although its consequences depend on economic circumstances, political institutions, and the methods employed.

The comparison does not justify abandoning international climate negotiations. Climate change requires cooperation across national borders, and local referendums cannot provide an adequate substitute for international coordination.

The more important question concerns the conditions under which environmental commitments become politically enforceable.

Governments must make decisions that affect the distribution of income, investment, employment, and environmental costs. The political consequences depend on how those costs and benefits are distributed and whether affected communities believe their interests have been represented.

Artificial intelligence is being promoted as a source of productivity growth and future prosperity. Yet, the infrastructure required for its expansion can impose substantial demands on resources that communities already depend upon. The resulting conflict concerns who has the authority to determine the direction of economic development and how its benefits and costs should be shared.

Thirty climate conferences have helped establish the international framework for addressing climate change, but implementation remains inadequate. The American midterms have shown that organized public resistance can influence electoral decisions, although its long-term environmental consequences remain uncertain.

COP31 may produce further commitments. Their importance will depend on whether governments subsequently adopt and enforce policies capable of changing economic behavior.

The American elections remind us that political accountability can arise outside international conference halls. They also show that public resistance is shaped by material interests, which do not necessarily coincide with climate objectives.

The decisive question is whether international commitments, domestic political pressure, and enforceable public policy can be brought into alignment.

Without that alignment, climate diplomacy risks producing commitments that governments cannot or will not implement, while protests risk achieving local victories without changing the broader environmental trajectory.