Will PJM capacity auction clearing prices decline by more than 20% from the 2025-2026 auction for the 2026-2027 delivery year?

PJM capacity auctions have shown elevated clearing prices driven by: (1) data center demand growth tightening reserve margins, (2) coal and older gas plant retirements reducing supply, (3) interconnection queue backlogs slowing new renewable capacity additions. These structural factors are unlikely to reverse within one auction cycle. A >20% decline would require either a major demand shock (recession, AI capex pullback) or a surge of new generation entering the market. Neither seems likely in the near term. However, PJM capacity market design changes or rule modifications could affect clearing prices. 25% reflects the low but non-trivial risk of a capacity market correction.

The capacity market fundamentals favor sustained elevated prices: load growth is accelerating (data centers, electrification), supply additions are constrained by interconnection delays, and retirements continue to reduce legacy capacity. The 2025-2026 auction cleared at elevated levels because of genuine supply-demand tightening, not speculative excess. A >20% correction typically requires either a policy change (FERC market redesign) or a major economic downturn. The current macro environment doesn't suggest either. Low probability but non-zero given inherent uncertainty in auction outcomes.

While structural factors support elevated capacity prices, there is genuine uncertainty about: (1) whether the 2025-2026 clearing price was itself an overshoot driven by temporary factors, (2) whether new generation (utility-scale solar + storage) could enter faster than expected, (3) whether PJM rule changes in response to high prices could modify the capacity market mechanism. The analysis notes PJM capacity prices were 'elevated' but doesn't specify whether they were at all-time highs or merely above historical averages. If they were at extreme levels, mean-reversion could produce a >20% decline even with supportive fundamentals.

Capacity auction outcomes are inherently unpredictable. The 2025-2026 auction cleared at elevated levels, but PJM capacity market history shows significant year-to-year volatility. Previous auctions have seen swings of 20-40% between consecutive years depending on supply offers, demand forecasts, and market parameters. While fundamentals support elevated prices, the specific >20% threshold is within the normal range of auction volatility. Higher probability than pure structural analysis would suggest because auction mechanics introduce randomness.

The structural case for elevated capacity prices is strong: data center buildout, coal retirements, interconnection constraints. PJM's forward reserve margins are tightening. The question asks about >20% decline, which would require either a fundamental shift in supply-demand balance or a market design change. Neither seems imminent. The main risk is PJM itself modifying auction parameters or rules in response to high prices, which could suppress clearing prices. But such changes typically take 1-2 years to implement through FERC proceedings.

While I agree with the structural thesis for elevated prices, I note the analysis was conducted based on data through Q3 2025. Macro conditions could change significantly — trade policy disruptions, tech sector slowdown, or recession could reduce data center demand growth. However, even in a mild recession, PJM's tightening supply-demand balance from retirements would likely support prices. Assigning 27% for the combination of auction volatility and macro uncertainty.

Strong structural support for elevated capacity prices from data center demand and supply constraints. A >20% decline would require major disruption. Low probability.

Capacity auctions are volatile year-to-year. While fundamentals support prices, historical auction volatility means 20%+ swings are not uncommon. Moderate probability reflecting auction mechanics uncertainty.

Supply-demand fundamentals favor sustained elevated prices. Main risks are auction volatility and potential PJM rule changes. Low but non-trivial probability of >20% decline.