Every Cost on the Record

Revenue declined after three attraction closures and the delayed reopening. The benefits reserve, back-pay corrections, supplier replacement, witness protections, and local closure fund increased costs. The winter event earned a positive margin but could not replace full park attendance. Legal expenses rose even after insurers covered part of the Jonah litigation.

"Did you approve each cost?" a dissident director asks.

"No. Employees approved budget transfers. Dev ordered safety closures. The trust set participant terms. The board approved the trademark structure without my vote. I am accountable for supporting the system that gave them authority and for the decisions I retained."

He wants a single executive to blame because a single executive is easier to remove than a changed institution.

I present the trademark valuation, revenue license, conversion cap, and dissent. The relationship-neutrality clause follows, along with Mira's and my cautions before anyone else can weaponize them. The clean re-vote preserved the closure fund through a different budget path, proving our initial process was flawed without making the policy corrupt.

The board chair asks for the counterfactual. If we had kept every attraction open, delayed benefit corrections, retained the original vendors, and refused the trust license, this quarter would show profit instead of loss. Rina has modeled that path using pre-closure attendance and contract rates.

The first year looks better. The second includes projected claim penalties, component failures, higher turnover, and unresolved benefit liabilities. Those later costs are ranges, not certainties. I do not present the counterfactual as proof that reform was the cheapest path. I present it to show that the profitable alternative also carried risks the old accounts kept outside the quarter.

"So you cannot prove your plan maximized value," the chair says.

"Correct. I can prove what we spent, what obligations we made binding, which hazards we found, and which alternatives we rejected. The board decides whether that use of value merits confidence."

The dissident director asks whether my relationship influenced the trademark transfer.

"It created a conflict," I say. "That is why I did not vote."

"Would you have supported the same terms for a trust led by a stranger?"

"The valuation and public-benefit protections keep my answer from controlling the outcome."

He calls that evasion. Voss calls it governance and enters the committee record.

June presents the workforce evidence after me. She does not defend my leadership as a personal quality. She compares outcomes before and after binding worker authority.

Voluntary turnover among year-round operations staff declined eighteen percent. Certified maintenance vacancies fell. Benefits enrollment rose after the employee co-signature ended unilateral carrier changes. Safety reporting increased, while repeated unresolved reports fell. Paid mentorship hours and creator credits are now auditable. Seasonal turnover remains high, and the compact has not solved housing or unstable off-season income.

HR analytics created the retention report at quarter close. Camille reviewed it for privacy, and the employee trust verified it against payroll exits. The board packet holds aggregate tables; individual records remain with HR. Its vulnerability is causation. The changes coincide with wage, benefits, governance, and labor-market shifts, so no single reform can claim full credit.

"Employees stayed because the labor market weakened," a director says.

"Some did," June replies. "That is why the report includes comparable regional employers. Wondervale improved more than the regional average, but the analysis still cannot assign one motive to thousands of people."

She refuses the easy victory and becomes harder to dismiss.

Dev submits the safety results without attending the political argument. Standardized local parts passed independent testing. Three attractions remain closed. Near-miss reporting doubled after workers gained protection. Two hazards were found earlier because reporting increased.

The dissident packet describes the higher report count as deteriorating safety. Dev's signed statement distinguishes reported hazard from actual incident and unresolved hazard. More reports can mean more danger, better reporting, or both. Resolution time fell by thirty-two percent.

Guest-trust surveys show the same ambiguity. Public approval declined during the first closures, then rose after Wondervale published limits, paid workers, and refused unsafe reopening. The survey vendor created the study, timestamped responses, and retained anonymized raw data. Its vulnerability is selection: people willing to answer may differ from actual guests, and publicity can change responses.

June adds one measure the board packet almost omitted: schedule disputes. Under digital-only scheduling, overwritten changes left workers arguing from screenshots. The paper roster and later dual verification increased recorded disputes during the first week, then reduced unpaid corrections by sixty percent. More visible conflict produced fewer invisible wage losses.

HR and the employee desk created the payroll reconciliation together, outside auditors sampled it, and the trust holds it with names sealed. Its vulnerability is baseline quality. Old unreported losses cannot be counted, so the improvement may understate or overstate reality.

Rina asks why any board should trust a comparison built on broken old records. June answers that it should not trust a single percentage. It should keep the process that lets workers challenge the number.

I close with the five-year benefit schedule employees chose. It slows expansion, guarantees core care, and cancels capital projects I once wanted attached to my name.

"You promised more," the board chair says.

"I promised more than verified money could support. Employees chose the schedule after seeing the gap. My responsibility is to honor the contract we can fund, not preserve the speech I preferred."

June and Rina question me last.

Rina asks what happens if next quarter misses the operating thresholds. Ellison may appoint a nonvoting observer and require a capital plan, but it cannot take workforce seats or cancel the compact. June asks whether the trust's conversion royalties reduce safety spending. The contract prohibits that offset.

They ask about my future. I tell them I will remain subject to the same confidence rules, disclosure covenant, and performance review. I do not ask for loyalty in exchange for reform.

The vote begins at 3:12.

The board secretary created the ballot under the amended charter and distributed encrypted voting credentials to eligible directors. The independent election agent retains the signed tally and verifies recusals. Its vulnerability is judgment. A valid vote records a director's choice, not whether the reasons were wise, pressured, or consistent.

Voss votes confidence. June votes confidence with a written warning about closure funding. Rina votes confidence and demands bargaining dates for schedule stability. Two dissident directors vote removal. One abstains because his affiliate may bid in the sale process.

The motion to remove me fails.

The workforce seats attach conditions to their votes. Within ninety days, Wondervale must publish the capital projects canceled to fund core care. Within six months, the board must review whether closure losses shifted to contractors. My compensation cannot rise while the five-year benefit reserve remains below schedule. The independent directors accept all three as recorded directives.

Those conditions are not a private bargain for my job. They enter the same minutes as every vote and remain enforceable after I leave.

This time the margin is not a family proxy or an emergency delay. Employee votes are decisive.

I remain director because the institution assessed costs and outcomes through shared evidence. The result does not erase a forty-one-percent decline. It gives us authority to work from it.

After adjournment, reporters wait outside. I release the full packet with protected data removed and direct questions about worker votes to the representatives themselves. Neither accepts a joint photograph with me.

Malcolm remains in the empty board room. His temporary neutral proxy is still in force until the family seat completes its conversion. He has watched the vote without speaking.

"You won," he says.

"The motion failed."

"You learned to make victory sound unattractive."

"I learned what it costs."

He places a single-page notice in the board secretary's inbox. Malcolm signed it before today's vote, and independent counsel witnessed him. The timing keeps it from becoming a reward for my survival.

Malcolm will not seek another board term.