Environmental projects end. Infrastructure can't

New York Climate Week has just concluded, and one shift in the environmental conversation was difficult to miss. Nature is increasingly being discussed not simply as something to protect, but as infrastructure: watersheds that provide water security, wetlands that reduce flood risk, forests that protect communities and coastlines that support economies.

That shift matters. Treating environmental systems as infrastructure can change how governments plan, how investors allocate capital and how environmental outcomes are valued. But taking the infrastructure analogy seriously also raises a less comfortable question: what happens after the project ends?

A bridge may begin as a capital project, but responsibility for it does not disappear when construction crews leave. Roads have authorities responsible for their upkeep; water systems have operators; power networks have maintenance schedules, regulators and revenue arrangements. The institutions may change over time, but the expectation that someone remains responsible for performance does not.

Environmental interventions are often organised differently. A wetland restoration programme may have a grant period, a watershed project a funding cycle, a plastic interception system a pilot phase and a resilience programme a defined set of deliverables. Each may have a beginning, a budget and a completion date.

The environmental service we expect from them rarely does.

That mismatch deserves more attention: we often finance environmental interventions episodically while expecting environmental services continuously.

When a project becomes infrastructure

Consider New York City's drinking water. Rather than relying solely on filtration infrastructure, the city has spent decades protecting the Catskill and Delaware watersheds that supply most of its drinking water. The familiar lesson is that a healthy watershed can perform an enormously valuable infrastructure function. What interests me here, however, is what had to develop around it.

The watershed arrangements that emerged in the 1990s brought New York City, watershed communities, regulators and other institutions into a continuing system for protecting water quality. Agreements, monitoring, land-management programmes, regulatory authority and relationships with landowners have had to persist through political, ecological and economic change.

The watershed performs a natural infrastructure function. But the water and forests alone do not constitute the infrastructure system. The arrangements that hold people and institutions responsible for performance are part of it too.

Toronto's Port Lands offers a rather different example. A major public investment has naturalised the mouth of the Don River, creating a new river valley that provides flood protection while enabling the redevelopment of a substantial part of Toronto's waterfront. The result is a striking example of the economic and social value environmental infrastructure can create.

Yet the project's completion reveals something less visible and perhaps equally important. Completed assets have to move into operating budgets. Sediment still has to be removed. Landscapes and infrastructure must be maintained. Toronto is now confronting precisely these questions as the Port Lands move from a transformative project into an enduring part of the city's infrastructure.

The celebrated moment is the new river. The harder test comes in year five, year fifteen and year thirty.

The temporal problem

It is tempting to describe this simply as maintenance. I think the issue is larger.

Environmental systems are dynamic. Rivers move; sediment accumulates; vegetation changes; and ecosystems respond to shifting temperatures, rainfall and human pressures. Political administrations turn over; communities change; and the economic uses surrounding an environmental asset evolve. The conditions under which an intervention was originally designed may not be the conditions under which it eventually has to perform.

Long-term responsibility therefore means more than maintaining what was originally built or restored. It requires the capacity to observe performance, learn from changing conditions and adapt.

This is where environmental infrastructure begins to look different from a succession of environmental projects. The distinction is not whether one involves nature and the other concrete, or whether private capital has been mobilised. It is temporal: a project can finish successfully when its agreed activities have been completed; infrastructure is expected to continue delivering a service.

If that is right, then one of the central design questions for environmental infrastructure is not merely how to finance an intervention, but how to make responsibility for its performance persist beyond the intervention itself.

Beyond the capital stack

Environmental finance has become considerably more sophisticated at assembling capital around projects. Grants can absorb early-stage risk, development banks can provide concessional capital, guarantees can change risk allocation, private investors can finance suitable cash flows and outcome payments can reward verified performance.

These innovations matter, and much of my work has involved thinking about how to combine them more effectively. But a well-designed capital stack does not necessarily produce a durable operating model.

The distinction is familiar in conventional infrastructure. Financing construction does not remove the need for operators, maintenance, monitoring and eventual replacement. Infrastructure finance has always had to confront both capital expenditure and what happens afterwards.

Environmental finance has sometimes been less explicit about the second half of that equation, perhaps because many of the services involved do not produce straightforward revenues. Monitoring a protected marine area, controlling invasive species in a restored wetland, maintaining relationships with upstream landowners or collecting environmental data may all be essential to performance without generating the revenues needed to support them.

That does not make them peripheral costs. If the environmental service depends upon them, they are part of the infrastructure proposition.

Continuity without permanence

In my previous essay for this series, I suggested that recognising nature as infrastructure is only the beginning. Environmental infrastructure also requires an institutional architecture around the asset: public authority, operating capacity, revenue arrangements, procurement, data, accountability, community relationships and long-term stewardship.

The question now is how those functions persist.

This does not require the same organisation to remain responsible forever. Conventional infrastructure rarely enjoys that kind of permanence. Governments change, operators change, concessions expire and technologies are replaced. What matters is that responsibility can be transferred without the underlying service becoming nobody's responsibility.

For environmental infrastructure, that continuity might come through a public authority with a long-term mandate, a utility or operator funded to maintain a service, community institutions with recognised authority, dedicated revenues, monitoring systems that trigger intervention, or contractual arrangements establishing what happens when performance deteriorates. Often several will be required.

The distinction I find increasingly useful, then, is between financing an environmental project and financing the conditions under which an environmental service can continue.

The morning after financial close

This distinction may become more important as nature, water and resilience move further into mainstream finance. Considerable effort is going into making environmental projects investable, developing project pipelines and mobilising institutional capital.

But financial close is a milestone, not an operating model.

For environmental infrastructure, due diligence might therefore extend beyond the familiar questions of return, risk and measurable impact. It should also examine who has authority to intervene if performance deteriorates, where operating resources will come from, who maintains the information needed to assess performance, and how responsibility survives changes in environmental conditions, funding, investors or political administrations.

These questions may attract less attention than a new financing instrument. They may nevertheless determine whether that instrument ultimately financed infrastructure or simply another project.

This suggests a practical test for the emerging Environmental Infrastructure conversation. Before calling an environmental intervention infrastructure, we should be able to identify not only the service it provides and how its creation will be financed, but how responsibility for its performance will survive the project that created it.

Projects have completion dates. Environmental services do not.

If nature really is infrastructure, financing the beginning is only part of the job.

Monty Simus is Senior Advisor for Blue Finance at The Ocean Cleanup and leads illuminem's Environmental Infrastructure initiative. He is also a PhD researcher at the University of Birmingham's Treatied Spaces Research Group. His work sits at the intersection of environmental infrastructure, innovative finance and governance, with professional experience across Asia, the Pacific and North America. His current writing explores how environmental outcomes can be financed, governed and sustained at scale, while his academic research examines Indigenous jurisdiction and resource politics in Bristol Bay, Alaska.

illuminem Voices is a democratic space presenting the opinions of leading Sustainability Thought Leaders, their views do not necessarily represent those of illuminem.

The world needs sustainability knowledge. At illuminem, no interest group or shareholder can influence our work. Thank you for supporting our mission to make high-quality and independent sustainability information free for all. Every contribution helps. Thank you for donating today.