PJM Interconnection's May 6 white paper sets out three distinct approaches to curtail what the system operator calls a "credibility trap" - a dynamic in which political responses to tight market conditions erode investors' confidence in scarcity-driven price signals. Mizuho's analysis of the white paper concludes that the differential reliability concept, identified as Path B, is the most politically feasible path, though it brings significant operational questions.
Path A - Mandatory Long-Term Hedging
Under Path A, PJM would require most load to secure long-term hedges or capacity contracts prior to participating in the existing capacity auction. That change would reduce the size of the residual spot auction where exiting generators currently clear and recover net cost of new entry. PJM's timeline for enacting Path A's hedging reforms is between 2026 and 2029, according to the white paper.
Path B - Differential Reliability
Path B would create a system of prioritized service during emergency periods based on whether loads pay for guaranteed supply. In this framework, customers that fund their own supply would receive priority access. The white paper outlines that data centers or other load-seeking guaranteed service could be required to build on-site generation, enter long-term bilateral power agreements, or purchase a reliability tier that provides prioritized service. Mizuho characterizes this option as the most likely to advance politically, while noting the approach could be difficult to operate without more advanced, large-scale dispatch technologies to manage the grid.
Path C - Shrinking the Capacity Market
Path C envisages redesigning the market so the capacity construct functions primarily as a backstop, shifting the principal investment signal toward the energy and ancillary services markets. PJM indicated any move to this model would not begin before 2028. Mizuho describes Path C as the most radical of the three options and the least likely to be implemented.
The white paper explicitly considers arrangements for data centers to co-locate with generators or to contract directly for supply, a matter that all three pathways would need to accommodate.
Stakeholder Responses
During its first-quarter 2026 call, Constellation Energy management signaled a preference for Path A, expressing support for a competitive framework that co-optimizes energy and reserve markets and raising concerns that Path B's differential reliability could treat different loads disparately. Meanwhile, FirstEnergy registered opposition to Phase 2 of the proposed reliability backstop mechanism in its own first-quarter call, arguing that wires-only utilities should not assume commodity, load and forecasting risk.
Implications
- All three reform pathways would alter incentives and risk allocation between load, generators and transmission providers.
- Data centers and other large, inflexible loads face explicit choices under Path B about self-supply, contracting or paying for priority service.
- The timing differences among the paths - with implementation windows cited for 2026-2029 and a redesign not starting before 2028 for Path C - suggest a multi-year policy and market transition.