Every water shortage already has a payor. The check just gets written late — after the reservoir drops, after the intake fouls, after the permit stalls, after the neighbors start showing up at hearings.
That is the plain starting point for anyone budgeting for water shortage under climate change. It does not arrive as weather. It arrives as hauling contracts, fallowing payments, emergency leases, cooling retrofits, and a capital plan sized for a source that keeps producing less than it used to.
So the honest answer to who pays to create supply is: the same people who are already paying. Water utilities, cities and governments, corporations with basin exposure, and data centers. The question was never whether they pay. It is whether the check lands before the water is gone or after.
shortage is already in the budget
Shortage almost never appears as a line item called shortage. It hides inside line items that look like ordinary operations.
A drought surcharge is shortage. A leased acre-foot is shortage. A fallowing payment is shortage. A deeper well, a second intertie, a new intake, a desalination study, a closed-loop retrofit — all shortage, priced as capital and defended as prudence. The spending is real, most of it is justified, and almost none of it changes how much water the basin produces next year.
A shortage budget buys a bigger share of a shrinking supply. It does not grow the supply.
That is the whole gap. Everything in the shortage column moves existing water — around a system, between users, from one season to another, or out of a wastewater stream. Only one kind of spending changes the amount of water the land catches, holds, and releases late in the year.
the payor map
Four payors, one pattern. Each has a hard-dollar dependency on a specific basin. Each already writes a shortage check. Each could write a different one.
| payor | the dependency | the check already written, after | what the earlier check buys |
|---|---|---|---|
| Water utility | How much water arrives above the intake, when it arrives, and how dirty it is | Emergency purchases and leases, drought surcharges, post-fire turbidity and sediment handling, treatment and intake capital | Condition on the ground that decides yield — infiltration, wet meadows, floodplain storage, fuels work above the reservoir |
| City, county, or state government | Reliability for households and growth — and, on working rivers, freight movement and power-plant cooling | Hauled water, new wells and interties, connection moratoria, lighter barge loading and curtailed generation in low-water years | Late-season flow on the reach the local economy actually runs on, and a longer usable season for everything that reach carries |
| Corporation | A basin-level license to operate, plus a supply chain that irrigates | Contested permits and renewals, replenishment pledges bought one project at a time, input volatility, disclosure exposure | A durable, funded position in the basin the business actually draws from — not a portfolio of unrelated projects |
| Data center | Consumptive cooling in the basin the site was chosen in | Hookup negotiations, closed-loop retrofits that move the cost onto the power bill, delayed or blocked projects | Supply created in the same basin the community is asking about, on a timeline that outlasts the news cycle |
The same four buyers show up for conserved-water credits, and that is usually a reasonable first move. Who buys a water credit covers that purchase honestly — what it does, and where it leaks. This post is about the other check.
Two of these payors already have their math worked out in detail: water utility drought risk mitigation for the intake side, and data centers drink water for consumptive cooling.
the two checks
Put them side by side. Neither one is optional in the sense that matters — the shortage check gets written whether or not anyone decides to write it.
| the shortage check | the supply check | |
|---|---|---|
| When it lands | After the shortfall is measured | Before the next dry year is priced |
| What it buys | A larger share of what is left, or gallons at one tap | More water produced and held on the land |
| How long it lasts | One season, one order, one contract | Decades, if the place stays protected |
| Who else benefits | No one — someone else loses | Everyone drawing on the same reach |
| Whether it repeats | Yes, and it grows | Repeats less, if the place holds |
That last row is the part that gets missed in a rate case. A hauling contract in a bad year does not make the next bad year cheaper. Restored infiltration above the intake does.
None of this is an argument against building things. Desalination and reuse genuinely make freshwater at a tap where a coast or a wastewater stream is available, and efficiency is simply good management. But a plant does not raise late-season baseflow two hundred miles inland, and a conservation order does not add rain.
why no single payor funds it alone
A restored meadow above an intake produces water for the utility, the city downstream, the grower on the ditch, the plant that needs cooling water, and the data center that sited in the valley — all at once, for beneficiaries who never meet and never signed anything.
That is a joint product with diffuse beneficiaries, which is the oldest underfunding problem there is. Everyone depends on it, nobody owns the bill, so it stays underfunded until it fails. Where it has been funded — Denver's forest-to-faucet program, New York's Catskills deal, Santa Fe's watershed fee, the Rio Grande Water Fund — one large buyer wrote the check and the other beneficiaries rode free. That works until the buyer's budget cycle turns. The gap is not the idea; it is an instrument the other beneficiaries can hold. The Colorado River version of this is worked out in who pays to protect the Colorado River — the same structure applies to any basin with more than one serious water user, which is all of them.
The fix is not a bigger grant program. It is a shared funding position the beneficiaries can each hold a piece of, sized to what each one already loses when the water does not show up.
how a shared position actually works
This is where ensurance comes in, and it is worth being precise about what it is and is not.
The foundation is the natural asset — a named watershed, meadow, or reach whose stocks and flows can be measured and ensured. Nature-based solutions are the work on it (protect, restore, rewater). Ensurance is the mechanism that funds that underlying.
Payors fund that account directly, and hold a funding position that says so — a transferable certificate on the named place's account that funds present condition and stays on the books, not a grant expensed and forgotten. Two instruments do different jobs:
- Certificates are tied to one named place. Funding is direct: your money goes to the account for that watershed, and the position you hold names it.
- Coins are protocol-wide. Funding is indirect: trading activity routes proceeds into protection broadly, without picking a place.
For a payor with a specific dependency, the certificate is the relevant one. You are not buying an offset or a credit against a volume. You are funding the present condition of the land your supply comes from, and holding a position in that place rather than a receipt for a transaction that already closed.
Be clear about the boundary: ensurance does not create water. Soil, wetlands, floodplains, beaver, and forest structure create water — those are the natural assets, and the nature-based work on them. Ensurance is the mechanism that funds the underlying. Anyone who tells you a financial instrument makes rain is selling something else.
And our own stage is small. Fewer than three dozen certificates are live across the whole protocol today, with named-place accounts and groups running at modest volumes. That is early, and pretending otherwise would be the same move we are criticizing. The mechanism works and the dollars are small — which is exactly when a founding payor has leverage.
what a first conversation looks like
Not a portfolio, not a program. One dependency, one place.
- Name the dependency. The specific reach, aquifer, or reservoir your operation cannot function without.
- Name the place that produces it. Usually a headwaters area, a floodplain, or a recharge zone with a steward already working on it.
- Price what you already pay after. Pull the leases, the surcharges, the hauling, the retrofit, the delay. That number is the comparison, not zero.
- Find who else shares the place. The other payors on the same reach are your co-funders, not your competition for it.
If you run a water system, start at solutions for utilities. If you sit on the public side — a city, county, state agency, or port authority on a working river — start at solutions for governments.
If you already know which place your supply comes from and want to talk about funding it, start the conversation. One dependency, one place, one hour.
the series
This is the last of six posts on where new water actually comes from.
- what new water actually is
- conservation does not create water
- why european rivers run dry
- a plant makes gallons. a landscape makes rain.
- rewater the land
- who pays to create supply — you are here