When the State Abandons a Project Mid-Stream
There was a moment in 2023 that could have marked the beginning of something substantial. After decades of absence, winter chinook salmon returned to the McCloud River in Northern California, thanks to a partnership between the state government and the Winnemem Wintu Tribe, the ancestral guardians of that territory. The program was technically ambitious and politically unprecedented: it combined Indigenous ecological knowledge with state resources to reintroduce a critically endangered species. Finally, it seemed like a model of collaboration that could last.
But it didn’t last. California withdrew funding. The program found itself suspended in limbo, with fish already in the water but no operational structure to sustain what follows. As reported by the Los Angeles Times, Governor Gavin Newsom's strategy for salmon recovery now faces significant resistance precisely because it promised a transformation that was never built on solid foundations.
This is not news about fish. It is a postmortem of how decisions are made within structures where political momentum replaces institutional architecture.
The Issue Wasn't the Vision, but Who Supported It
The McCloud restoration initiative had all the ingredients of a project with genuine purpose: ecological urgency, community legitimacy, scientific backing, and political will. On paper, it was the type of program governments should replicate. The problem is that its viability depended on a fragile funding chain, tethered to annual budgetary decisions and, ultimately, to the political priority of a state executive with multiple simultaneous agendas.
When that support was withdrawn, no alternative mechanism was activated. There was no pre-identified consortium of private funders. There was no shared governance structure between the tribe and the state that could operate autonomously in the event of funding interruption. The program existed because someone at the top had pushed it, and when that person shifted their attention, the program ceased to exist in operational terms.
This pattern has a name in organizational theory, although here I am less interested in naming it than in describing its concrete consequences: when a system relies on the active will of its promoter to keep functioning, it is not a system; it is a promise. And promises do not restore ecosystems.
What the Winnemem Wintu Tribe faces today is not just a renewed historical injustice, although it certainly is that. It is the direct consequence of being incorporated into a structure that was never designed to operate without central oversight. They were asked to be partners but treated like executors of a mandate that could be canceled from above.
The Illusion of Visible Leadership as a Guarantee of Continuity
There is a tendency in public and corporate management that confuses the announcement of a program with its consolidation. A governor appearing in the photo of the salmon's historic reintroduction in the McCloud generates immediate political capital. The narrative is powerful: reconciliation, biodiversity, Indigenous rights, the future. But that political capital does not automatically reinvest in the infrastructure the program needs to survive electoral cycles or budget adjustments.
Leadership that builds on its own visibility tends to create programs that resemble it: brilliant at launch but fragile in everyday operation. This applies equally in a tech company as it does in a state government. The pattern is identical: the leader concentrates the project’s legitimacy in their figure, and when that figure moves on to another front, the project loses its center of gravity.
The opposite of this model is not the absence of leadership; it is leadership that actively works to make itself dispensable. That transfers real authority, not nominal delegation. That builds capacity in partners, not dependencies. That designs diversified funding mechanisms before announcing the program, not after the first one fails.
In the case of the McCloud, a mature structure would have included, from the beginning, a multi-year reserve fund, binding agreements with the tribe that could not be unilaterally canceled by the executive, and metrics for transferring capacities to the Winnemem Wintu community so that operational knowledge did not reside within the state’s organizational chart. None of this, according to available information, was built with that solidity.
What Remains When Funding Departs
The winter chinook salmon are now in the McCloud River. This cannot be undone, and it should not be downplayed: it is an ecological achievement with decades of debt still outstanding. But a fish reintroduced into a habitat without sustained monitoring, without a technically equipped community for follow-up, and without a budget to manage the ongoing threats it faces has survival odds very different from those that justified the initial effort.
This is what poorly designed governance systems frequently produce: interventions that generate dependency without creating capacity. An intervention occurs in a community's territory, a process is activated that cannot be halted midway without consequences, and then support is withdrawn as if the world can pause while budget issues in Sacramento are resolved.
For any organization, public or private, operating with external partners, communities, or territories, this story delivers a concrete operational lesson: the cost of exiting a program mid-life cycle almost always exceeds the cost of having funded it sustainably from the outset. Not only in reputational terms but also in terms of the institutional trust that gets destroyed and which, once shattered, takes generations to rebuild.
The Winnemem Wintu Tribe has spent decades striving to return the McCloud to what it once was. The state arrived too late, participated briefly, and left before finishing. The question that executives in any organization should ask themselves is not whether their programs are ambitious but whether they are built to operate when they are no longer watching.
The Mandate Structures Cannot Ignore
What the case of the McCloud River precisely exposes is the difference between a program and an institution. A program exists as long as someone funds and defends it. An institution exists because it was designed to outlive its creators.
Leaders who build programs are managing their agenda. Leaders who build institutions are managing the future of the communities that depend on them. The distinction is not philosophical: it has direct consequences on how funding is structured, how authority is distributed, how teams are formed, and what commitments are made before the first public announcement.
The model that produces real resilience is one where the original driver of the system works from day one to ensure their departure is not a crisis. Where transferring capacity to partners is a metric of success as important as the visible outcome of the program. Where leadership measures its effectiveness not by the headlines it generates, but by the solidity it leaves behind when it departs. That is the only standard that matters for those who build something meant to endure beyond their own mandate.
