Categories
|

Taiwan’s Nuclear Phase-out: (Almost) one year on

Taiwan’s Nuclear Phase-out: (Almost) one year on

By Chia-wei Chao, Yang Wei, and Benjamin Yang (Taiwan Climate Action Network)

Published at WISE/NIRS Nuclear Monitor Issue 937


At precisely 10 p.m. on May 17 2025, Taiwan’s last operational nuclear reactor, located at the Maanshan Nuclear Power Plant, was quietly taken offline. While there was no media spectacle or elaborate celebrations declaring triumph, this moment was undoubtedly a significant one. For a nation whose energy policy has been highly politically contentious for as long as, if not longer than, its democratisation, this marked a milestone in Taiwan’s energy transition and what seemed like the definitive end of an era.

When Taiwan’s ruling Democratic Progressive Party (DPP) initiated this energy transition in 2016 upon regaining power, the country’s electricity generation was dominated by coal, which accounted for 45% of total supply. Six nuclear reactors contributed around 14%, whereas wind and solar barely reached 1% combined in 2015.

Nine years later, under the government’s planned phase-out where nuclear reactors were decommissioned as their 40-year operating licenses expired, nuclear’s share dwindled to just 1.3% before the Maanshan shutdown. Coal’s share also declined to 35%, as renewables, primarily solar and wind, saw a significant jump to around 13%. Liquefied natural gas (LNG) now plays a central role in electricity generation, accounting for roughly 48% in 2025.

Renewable power generation is now 3.6 times higher than in 2015, with its share of the electricity mix reaching 13%. Yet compared with the original targets of 20% and 61.7TWh, the achievement rate stands at only about 60%. The main reason is the lag in ground-mounted solar development, which has left the original 20GW solar target at just 15GW.

Had the original target for cutting coal use in the power system been met, Taiwan could have reduced national coal consumption by 34%. Instead, the reduction so far is only 20%. The growing reliance on natural gas to fill the power gap had little effect on costs before 2022, when gas prices were relatively low. Since then, however, Russia’s invasion of Ukraine and the crisis in the Strait of Hormuz have driven up gas costs, underlining the national security risks of this approach. It has also fuelled renewed questions over Taiwan’s nuclear-free homeland policy.

This transition, while still incomplete, reflects a profound reimagining of the nation’s energy landscape, with implications well beyond the energy system itself, on sovereignty, economic competitiveness, and resilience to both the climate crisis and geopolitical shocks.

It has not been a smooth one either. Since 2018, there have been a whopping total of six energy-related nationwide referendums. The sixth one, held in August 2025 on a proposal to restart the Maanshan nuclear power plant, was defeated as the votes in favour fell short of reaching the 5 million threshold.

As Taiwan’s nuclear phase-out approaches its one-year anniversary, however, how much longer the country’s energy system will remain nuclear-free remains a question as contentious and uncertain as ever.

From Nuclear-free homeland to the Three Principles: Driving forces behind the Lai administration’s nuclear restart plan

Even well before the Maanshan referendum took place last August, public opinion on the government’s nuclear phase-out, or what is more often dubbed “nuclear-free homeland” policy, had begun to turn.

Among the popular narratives favouring a return to nuclear, anxieties surrounding the need to power Taiwan’s world-leading artificial intelligence (AI) and semiconductor industries, often referred to as the “sacred mountain” protecting the nation, are perhaps the most prominent. The European Union’s framing of nuclear as a key contributor to decarbonisation has had a spillover effect as well, as nuclear proponents posit it as a stabler means to combat the climate crisis compared to intermittent renewables and as an alternative to using LNG as a bridge fuel.

Energy security challenges highlighted by the looming threat of a Chinese maritime blockade and the present Strait of Hormuz crisis, plus several bottlenecks in renewable energy development in recent years, have also prompted some DPP supporters to reconsider their historically anti-nuclear stance.

All these factors have translated into an apparent willingness among voters across party lines to accept nuclear power as an option in Taiwan’s future energy mix, according to recent opinion polls.

In what has been the government’s most emphatic declaration yet, President Lai Ching-te announced in March 2026 to a room of entrepreneurs that the state-owned utility, Taipower, is set to submit its plans for restarting the Maanshan and Kuosheng plants, the latter of which was decommissioned in 2023, to the Nuclear Safety Commission (NSC) for formal review, following the green light from the Ministry of Economic Affairs (MOEA).

Lai cited rising electricity demand driven by AI and data centre developments, new decarbonisation requirements such as the European Union’s Carbon Border Adjustment Mechanism (CBAM), and the ever-evolving geopolitical risks as factors informing this decision. He also emphasised that any restart would hinge upon three principles: nuclear safety, solutions for nuclear waste management, and public consensus.

What each of these three principles exactly entails, however, leaves room for interpretation: supporters of the restarts see them as hurdles to clear, while opponents take them as evidence of just how difficult and costly it is to reconsider nuclear.

Take nuclear waste management for instance, Taiwan has yet to find potential sites for final disposal repository facilities for either low-level or high-level radioactive waste, nor are there laws for siting a final disposal site for high-level waste. Opponents would contend that without such regulatory framework, let alone any clear pathway or timeline to siting a disposal repository, this condition will not be met; others have suggested that completing dry interim storage facilities that enable the spent fuel rods to be moved out of on-site cooling pools, which both the Maanshan and Kuosheng plants are in the process of doing, checks the box.

What constitutes public consensus is another thorny matter. The NSC plans to hold two information sessions in Hengchun, where Maanshan is located, with an aim of bringing local government authorities, representatives, and residents together to solicit feedback on the restart plan. It is unclear if these sessions would offer any indication of a consensus, but attempts to interpret them as such would not go unchallenged. Polls show that although around 60% of the public support an increase in the share of nuclear power, the proportion of those that consider nuclear power to be dangerous and those who cannot accept having a nuclear power plant in their own village or neighbourhood both exceed 60% as well.

More notably, only a quarter of the public has a correct understanding of the percentage of nuclear power in Taiwan’s total power generation, indicating the need for increased information integrity to serve as the basis for building any type of consensus.

That leaves us with nuclear safety, which is what the NSC’s ongoing reviews are set out to do. Taipower’s restart plans that Lai mentioned only cover basic information such as the current status of the sites and units and schemes on workforce allocation and training, facility inspection and maintenance, and quality verification and auditing. By law, Taipower will have to submit at least five more documents for the NSC’s review prior to any potential restarts:

  1. Integrated Aging Assessment and Aging Management: Review the plant’s past experience to conduct aging management assessments on systems, structures, and components (SSCs) and implement aging management programs.
  2. Time-Limited Aging Analyses (TLAAs): Conduct analyses and aging effect assessments for essential systems, structures, and components to evaluate whether they can maintain their normal functions during the extended operation period.
  3. Amendments to the Final Safety Analysis Report (FSAR) and Technical Specifications (TS): Revise the Final Safety Analysis Report and Technical Specifications based on the overall assessment results.
  4. Assessment and Verification of Radiation-Related Issues: Evaluate the radiation impact of restarting and extending operations on the public, workers, and the environment, as well as the management of radioactive waste.
  5. Seismic Safety Assessment Description: Evaluate the original seismic design of the plant and the enhanced seismic safety measures to be implemented during continued operation.
  6. Other documents designated by the competent authority.

The NSC has not given a timeline for when it will be able to complete these reviews, but MOEA has indicated to legislators that Maanshan could be cleared for restart by 2028 the earliest if all goes smoothly, citing the average review time of license renewal applications abroad.

But given the Maanshan plant’s history of various incidents and violations, ranging from fires to blackout events, and that it is located next to the Hengchun Fault, whether these reviews and subsequent upgrades can guarantee system safety, even with reinforced seismic resistance values, remains unknown.

A solution for energy resilience?

While the government and pro-nuclear opinion leaders have posited energy security as a key reason for at least considering restarting Taiwan’s nuclear plants, it is worth scrutinising how this argument runs and whether it holds up.

In recent years, Taiwan has faced escalating pressure from the United States to retain nuclear energy as a core component of its power grid. This “strategic urging” is primarily driven by Washington’s concerns over regional security and the stability of the global semiconductor supply chain.

Through the American Institute in Taiwan (AIT) and various security think tanks, U.S. officials have argued that Taiwan’s heavy reliance on LNG constitutes a critical defense weakness, as it is highly vulnerable to maritime blockade. From this perspective, nuclear power is framed as an essential “strategic buffer” that enhances national security by providing a carbon-free, on-site energy source capable of sustaining the island’s economy in the event of such crises.

However, when contrasted with the harsh realities of modern warfare, this U.S.-backed definition of “resilience” appears flawed. True energy resilience is defined not by gross generation capacity, but by a system’s flexibility and its ability to recover rapidly from extreme shocks. The war in Ukraine has shown that we can no longer assume nuclear plants will be safe during a conflict, with sites like Zaporizhzhia being weaponised for “nuclear blackmail.” An aggressor does not need to strike a reactor directly to create a crisis; simply disrupting external cooling or power lines to trigger a potential meltdown deters intervention through the threat of disaster.

Similar threats against the Bushehr and Barakah plants in the current Middle East conflict confirm that nuclear facilities are high-value strategic targets rather than protective shields. For Taiwan, restarting aging reactors is akin to “drinking poison to quench thirst.” Rather than providing security, these rigid, centralised assets become ticking time bombs that impose a massive defense burden and the risk of a radiological catastrophe in a conflict scenario.

An alternative pathway forward

Given the intense pressure of limited resources and a narrowing window of time, Taiwan must be precise in how it allocates its energy investments. Hoping for a nuclear restart may seem like a quick fix, but it fails to address the core vulnerability of being an island nation. Re-investing in nuclear diverts capital and resources away from the more cost-effective and ready-to-deploy renewable energy and storage solutions that help decentralise the grid and ensure reliable, localised power in times of crisis.

The European Union’s experience with the REPowerEU plan offers a compelling blueprint. Following the energy crisis triggered by the Russian invasion of Ukraine, countries like Denmark and Germany have demonstrated that integrating high levels of wind and solar power, supported by advanced battery storage and smart grid technology, creates a system that is far harder for an adversary to disable than centralised power plants. In the United Kingdom, aggressive investment in domestic offshore wind is now viewed as a primary defense against geopolitical volatility. These nations are moving toward models of energy democracy, where localised microgrids and distributed production ensure that even if one part of the network is compromised, the rest of the country can continue to function.

Ultimately, a centralised nuclear asset in a war zone is not a reliable source of power, but a high-stakes liability that deepens national insecurity. The way forward for Taiwan lies in a rapid shift toward a distributed and decentralised energy system. To achieve true resilience, national security, and meet the net-zero target, the strategic focus must stay on accelerating renewable energy development alongside enhanced grid infrastructure and energy storage.

Pursuing nuclear restarts incurs a heavy opportunity cost, as the massive capital and political energy required for nuclear power drain resources away from more effective renewable solutions. By localising energy production and reducing dependency on imported fuels, Taiwan can build a grid that is not only low-carbon but strategically robust and truly resilient against external pressure.