Externalities, Regulation, and the Future of US Nuclear Power Safety

Energy systems, while crucial for societal development, often come with non-market issues known as externalities. These are costs or benefits incurred by individuals who did not choose to incur them and are not exchanged in the marketplace. For instance, coal power plants pollute the air with sulfur and PM10, wind farms generate noise and can harm birds, and nuclear reactors, though rarely, can release radiation. Assessing energy systems purely on an economic basis without considering these externalities provides an incomplete picture.

Externalities distort the calculation of social welfare, which aims to maximize consumer surplus and ensure optimal resource allocation for societal benefit. When externalities are not factored into the supply curve, the perceived cost of production is artificially low. This leads to an overproduction of goods or services with negative externalities, resulting in a "deadweight loss" – a total loss of social welfare.

Beyond deadweight loss, externalities also create an involuntary wealth transfer. Those who benefit from the cheap production (e.g., consumers of electricity) do not bear the full cost, while others (e.g., those living near a polluting plant) are forced to pay for the harm without receiving compensation. This raises ethical concerns about fairness and equitable distribution of costs.

Addressing Externalities

Several approaches exist to address externalities:

1. Caldor-Hicks Approach (Compensation): This involves transferring value from "winners" (those benefiting from cheap electricity) to "losers" (those harmed by externalities). However, this approach is problematic due to:

   • Difficulty in Valuation: Assigning a dollar value to non-market losses like health impacts or environmental damage is challenging. Price reflects scarcity, not utility, making it hard to equate monetary compensation with lost utility.
   • Non-Generalizability: Monetary compensation has different utility for individuals with varying wealth. A sum that might compensate one person adequately could be insignificant to another.
   • Non-Marketable Losses: Some losses, like human life, are inherently non-marketable, making monetary compensation impossible or ethically fraught.

2. Pigouvian Tax (Economist Solution): This involves imposing a tax on activities that generate negative externalities. The tax aims to raise the supply cost to reflect the true social cost, thereby reducing the quantity produced to the socially optimal level and eliminating deadweight loss. While effective in optimizing social welfare, it doesn't inherently solve the wealth reallocation problem. The revenue from such taxes is often used for political ends, such as offsetting other regressive taxes or funding initiatives that benefit those most affected by the externality.

3. Regulation (Banning or Limiting): This approach seeks to prevent or minimize the existence of externalities through administrative rules and standards. Regulations, though not passed by Congress, have the force of law and are enforced by government agencies.

The Value of a Statistical Life (VSL)

A critical aspect of cost-benefit analysis for regulatory policies is estimating the harm caused by externalities, especially when human life is at stake. The concept of the "Value of a Statistical Life" (VSL) is used for this purpose. VSL represents the monetary value society places on reducing the risk of death.

Different regulatory agencies in the US (EPA, FDA, DOT, CPSC, FAA, NRC) use varying VSL figures, which have generally increased over time, reflecting societal affluence and a greater willingness to invest in safety. However, the Nuclear Regulatory Commission (NRC) stands out. Instead of a direct VSL, the NRC uses a value of $2,000 per unit REM (a measure of radiation exposure) avoided. This indirect approach results in a significantly lower implicit VSL (around $3.2 million) compared to other agencies (which range from $15 to $20 million). This discrepancy can lead to different regulatory outcomes, as illustrated by the post-Fukushima decision regarding filtered vents in boiling water reactors. European regulators adopted these relatively inexpensive safety upgrades, while the NRC, using its lower VSL, deemed them not cost-effective for US reactors.

The Role and Challenges of Regulation

Regulation is a set of administrative rules and standards that governments implement to minimize externalities to an acceptable level. While expensive (costing roughly 50% of the US federal budget and an average of $15,500 per household annually in taxes), regulation is crucial for maintaining modern living standards, ensuring product safety, and protecting the environment.

For the nuclear industry, regulation is particularly impactful. Licensing a new reactor design can cost up to $1 billion, and the overall cost of a nuclear power plant can reach $10-20 billion. This significant financial stake means the industry often seeks to influence regulations.

Challenges in nuclear regulation include:

• Industry Influence: The high costs involved can incentivize industry lobbying to change regulations, as the cost of influencing policy is often a small fraction of the overall project cost.
• Uncertainty in Harm Assessment: It's difficult to precisely quantify the long-term effects of radiation or the exact safety levels of reactors, making cost-benefit calculations for regulatory changes subject to interpretation and political influence.

Evolution of Nuclear Regulation in the US

The nuclear age began with the Atomic Energy Commission (AEC), which acted as both promoter and regulator of nuclear technology. This inherent conflict of interest led to issues, such as the AEC prioritizing cost-saving measures over safety, as seen in the debate between Westinghouse's "core catcher" and General Electric's "Emergency Core Cooling System" (ECCS). Experiments revealed the limitations of ECCS, leading to a scandal and congressional intervention.

In 1974, due to its perceived conflict of interest and failure to comply with environmental laws, the AEC was dissolved and split into the Department of Energy (promoting nuclear power) and the independent Nuclear Regulatory Commission (NRC).

Characteristics of a Strong Regulator

Academic literature suggests a strong regulator should possess:

• Independent Operations: Not reliant on the regulated industry for inspections or oversight.
• Independent Finances: Not financially dependent on the regulated entities.
• Independent Sources of Information: Capable of conducting its own research and analysis.
• Legal Authority: Empowered to enforce standards and regulations.

The NRC, while generally considered a gold standard globally, faces challenges in fully embodying all these characteristics.

Financial Structure and Conflicts of Interest

The NRC is mandated by law to recover 100% of its operating expenses (with some exceptions) from the industry through fees. While these fees go to the US Treasury and Congress allocates the NRC's budget, preventing direct financial incentive for overcharging, this "fee-for-service" model can create a perception of the industry as a client. This can lead to industry expectations that the NRC will act in their favor, especially when former industry personnel join the NRC.

Despite these challenges, the NRC has historically done a good job, with American reactors being among the most robustly operated worldwide. However, the industry often criticizes the NRC for increasing costs and hindering competitiveness.

Recent Changes and Political Influence

The Trump administration introduced significant changes to the NRC, challenging its independence:

• Personnel Changes: Replaced commissioners with individuals more aligned with the White House's views, including a former head of regulatory affairs for a major nuclear utility.
• Executive Order 14300: Mandated the NRC to consider the promotion of nuclear power alongside health and safety, reintroducing the conflict of interest that led to the AEC's dissolution.
• Staff Reductions: Directed the NRC to reduce technical experts on the Advisory Committee on Reactor Safeguards to the statutory minimum and cut the overall workforce.
• Capped Review Times and Fees: Imposed an 18-month cap on license reviews and capped fees, potentially compromising the thoroughness and independence of reviews.

These changes suggest a shift towards a less independent NRC, potentially leading to a greater reliance on industry-provided information and a less rigorous regulatory process.

Reactor Licensing Process Evolution

Historically, nuclear power plants were licensed in a two-part process (Part 50): a construction permit followed by an operating license after completion. This allowed for flexibility but also introduced financial risk for investors if safety rules changed during construction, as happened after the Three Mile Island accident.

To address this, the industry pushed for the Combined Operating License (COL) under Part 52. This allows utilities to apply for a license to build and operate a plant simultaneously. Once granted, the COL cannot be easily withdrawn or modified, provided the plant is built as specified. This shifts public opposition to the early "greenfield" stage, effectively disenfranchising local voices once construction begins.

The COL process, however, proved problematic in practice. The Vogtle and Summer projects, for example, experienced numerous license amendments due to design changes during construction, leading to significant delays and cost overruns. This led the industry to blame the regulator for inflexibility and advocate for further changes, such as shrinking the emergency planning zone to the plant's fence line.

Probabilistic Risk Assessment (PRA) and the "Risk-Informed Approach"

The proposal to shrink emergency planning zones relies on Probabilistic Risk Assessment (PRA). Historically, safety was based on "defense in depth" – multiple independent safety layers to account for unforeseen events. PRA, in contrast, involves building complex models of reactor operation, assigning probabilities to potential events, and multiplying these small numbers to arrive at a very low probability of core damage.

PRA originated from a 1975 MIT study (WASH-1400) that estimated a core damage probability of once every 20,000 years per reactor. This report was criticized for relying on estimated probabilities due to a lack of real data. While initially resisted by the NRC, PRA gained traction in the 1980s as a tool for evaluating the cost-effectiveness of specific reactor modifications.

While PRA has been useful in improving operational reliability (e.g., through the Maintenance Rule, which significantly increased US reactor uptime), it has limitations:

• Model Incompleteness: PRA models are inherently incomplete, as they cannot account for all possible failure pathways or human errors. Major nuclear accidents have often occurred due to pathways not included in the original PRA models.
• Human Bias: The probabilities assigned in PRA are often human estimates, introducing bias.
• Misleading Numbers: PRA often produces extremely low probabilities (e.g., one core damage event every three billion years for NuScale reactors), which are not always credible and can be used to justify reduced safety measures.

The "risk-informed approach" uses PRA to justify dismantling redundancies based on model uncertainty, potentially leading to a less robust safety framework.

Enforcement and Transparency

The NRC has enforcement powers, including the ability to arrest individuals who fail to perform their duties. The Office of the Inspector General (OIG) investigates instances of misconduct, which occur regularly (100-200 per year), including falsified records, violations of procedures, and retaliation against whistleblowers. Such misconduct undermines the assumptions of PRA models, which rely on ideal plant operation and adherence to procedures.

The NRC's mission statement emphasizes public transparency, stating that "Nuclear regulation is the public's business, and it must be transacted publicly and candidly." However, there are concerns about proprietary information being withheld from the public, even when it's not sensitive. The Union of Concerned Scientists has highlighted instances where the NRC withheld documents, including requests for exemptions from fire safety regulations, suggesting a potential "capture" of the regulator by the industry.

Conclusion

Regulation imposes significant costs but is crucial for safety and, paradoxically, can improve industry profitability. While imperfect and subject to influence from interest groups, it has generally enhanced safety. The shift towards a "risk-informed approach" and the recent political interventions raise concerns about the future independence and effectiveness of nuclear regulation in the US.