Nature finance runs on a theory of change that almost nobody writes down. Because it is never written down, it is never audited.
Here it is, stated fairly: set a target, build a framework, require disclosure, harden disclosure into a mandate, let markets price the risk — and capital follows. Serious people staff every link, and the chain has produced real things: better ecological science, a shared vocabulary between ecologists and CFOs, and a defensible case that ecosystems are financially material. What it has not produced, at scale, is a transaction.
That gap is the whole argument. Policy produces commitments. Only a product produces a transaction.
the money was never the scarce part
Nature-based solutions drew roughly $220 billion in 2023. About $197 billion of that — roughly 90% — was public. Private finance was about $23 billion. Activities that directly damage nature drew roughly $7.3 trillion — more than thirty times as much. Meeting the Rio Conventions commitments would take nature-based solutions finance to about $571 billion a year by 2030, so the shortfall is real, and wider estimates of the biodiversity financing gap run higher still.
Read that as a diagnosis, not an outrage. The $7.3 trillion is not withheld from nature out of malice or ignorance. It is deployed, because it has somewhere to go: instruments with a return, a term, a counterparty, and an exit. Nature has targets, frameworks, disclosures, and reports. It has very few instruments. The scarce input was never capital, and it was never concern. It is product.
audit the chain: what environmental markets actually transmit
The useful question at each link is not "does this matter?" It is: what must be true here for a dollar to move at the next link? Call that transmission. Environmental markets are the places a nature-related claim gets a price and a buyer. Most of them are assembled from links that transmit information beautifully and money not at all.
| link | what it produces | what must be true to move a dollar | transmission |
|---|---|---|---|
| target | political intent, a deadline | someone must become obligated or paid to hit it | none on its own |
| framework | shared method and vocabulary | adoption must change a price or an obligation | weak |
| disclosure | a number | the number must reach someone who now owes money | breaks |
| mandate | an obligation on the discloser | the obligation must become a bid for a specific asset | usually breaks |
| priced risk | a wider spread, a higher discount rate | the price must fund the exposed asset | inverts |
| capital | allocation | an instrument must exist to allocate into | no instrument, no allocation |
Three of those breaks deserve naming plainly.
Disclosure produces a number, not a payor. A disclosure regime is an excellent census: it tells you which balance sheets sit on top of which ecosystems, and that is genuinely new information. But nobody owes anything on the strength of a number. Between "this company depends on that watershed" and "this company pays for it" sits an obligation disclosure does not create.
A mandate binds the discloser, not the asset. When a reporting duty hardens into law, it lands on the reporting entity and converts into legal, consultant, and data spend — real economic activity, none of it a bid for an acre. The cheapest route to compliance is usually to shed the exposure, not to fund the ecosystem underneath it.
Priced risk moves capital away from the exposed thing. This link does not fail to transmit; it transmits in reverse. A more accurate discount rate on a fragile watershed is a better description of a dying watershed, not a healthier one — and it steers money somewhere less exposed. Wider spreads can attract specialist capital; they still do not, by themselves, fund the exposed place. We argue that in full in pricing nature risk is a symptom, not a cure.
the links that do work
One link has a real transmission mechanism, and it is worth understanding why. A statutory obligation that names an obligated party, sets a quantum or a rate, and attaches a consequence for non-compliance can move dollars — when it also manufactures a unit a supplier can sell. England's biodiversity net gain regime and US wetland mitigation banking under Clean Water Act Section 404 do that: someone must buy a defined unit at a price. The English unit market is thin and enforcement capacity is uneven; what it proves is the buyer, not a deep, reliable clear. A stormwater fee or a wildfire-resilience code can move money too, but those are usually payments to a utility or mandated works, not a traded unit.
The honest claim is narrow. Policy is excellent at floors: prohibitions, designations, protected status, and the legal recognition that makes titled land protectable — alongside customary and communal tenure, and Indigenous stewardship that already holds much of the world's remaining biodiversity. Much of the most important protection in history arrived that way. Policy manufactures buyers only when it attaches a price to a prohibition. When it does, the product inherits every risk of its statute: demand stops at the statute's border, and price tracks enforcement intensity rather than ecological condition. We take that case seriously in demand by statute — it is the strongest case for the other side.
nature has to travel inside the instrument
Capital allocates against return, risk, duration, liquidity, and fit to mandate. Nothing on that list is a value judgment. An allocator who personally cares a great deal still cannot underwrite care — they underwrite cash flows, and they answer to a committee reading those columns, plus ticket size and benchmark fit.
Which means the protective outcome has to ride inside the instrument as a property, rather than arrive alongside it as a reason.
Capital doesn't need to love nature. It needs to be unable to make money without protecting it.
That is a structural claim, not a moral one. Compare the two designs:
| nature as a reason (bolted on) | nature as a property (baked in) |
|---|---|
| a screen ranks the holder of the asset | the cash flow depends on the condition of the asset |
| disclosure reports the exposure | every transaction routes proceeds to the asset |
| a credit fragments one service out of an ecosystem | the whole asset stays intact and gets paid |
| concessional capital covers the missing return, indefinitely | the return and the protection are the same cash flow — which still has to clear on its own price |
Bolted-on designs ask capital to change its mind. Baked-in designs let capital stay what it is — self-interested, yield-seeking, indifferent — and put the caring in the architecture. That is what regenerative finance, or ReFi, means when it is a design rather than a label; the money design itself is worked out in living money.
One guardrail, stated every time: the source asset is never collateral — not pledged, not fractionalized, not foreclosable. Capital gets the coupon. It never gets the forest. That is the difference between naturalizing finance and financializing nature, and it is load-bearing, not stylistic. The credit consequence is equally load-bearing: there is no foreclosure on the land. Recourse sits on the payment rail and the protocol's ability to keep routing proceeds, not on seizing the place. That is a deliberate cap on the claim, not an oversight.
what would falsify this
A theory of change that cannot be wrong is not a theory.
The policy-first chain would be vindicated by: a voluntary framework producing an obligated payor without a statute behind it; disclosure adoption followed by measurable capital flowing into nature-based assets rather than merely away from exposed ones; a nature unit whose price tracks ecological condition instead of enforcement intensity; or one asset class that reached scale by mandate without the mandate itself manufacturing the unit, the buyer, and the supplier. Cap-and-trade and similar statute-created quantity markets are the obvious counterexamples; we take them in markets before mandates.
Our side fails if nature-linked instruments cannot produce a market-clearing return without permanent concessional support — at which point the protective property is decoration, not structure. That test is the subject of the subsidy that proves the product. We watch our own failure condition more closely than theirs, because it is the one we can cause.
the replacement, and where we actually stand
Our attempt to put the property in the product is ensurance. Three plain parts:
- coins — general ensurance. A fee on ordinary trades routes to protection. That fee is volume-linked and the weakest of the three rails: it is not a contracted coupon. It still has an address, and it is non-discretionary once the trade happens.
- certificates — specific ensurance, issued one-to-one against a named natural asset. Direct funding to a place, with condition on the record.
- proceeds — the routing layer sending value to the agents that represent a place, a people, or a purpose. Not a pledge to spend later — a rail that spends on the transaction.
What this is not: a claim of arrival. We have live agents, coins, and certificates, and small volumes. The valuation layer — stocks, flows, condition, and the gap between what a place produces and what it earns — runs on real parcels. The coupon that would make this boring enough for institutional capital is still being manufactured. Anyone claiming otherwise about any nature product, ours included, is describing a roadmap.
The difference we will defend is the direction of the causal arrow. Not: convince capital to want nature. Instead: build the instrument capital already wants, and make protection the thing it cannot get the yield without. A municipal reader designing a statute should ask whether it manufactures a unit and a supplier, not only a duty.
Talk through a nature product →
the series
This is part of a series on why nature gets funded as the outcome of ordinary products, not the reason for them.
- capital doesn't invest for nature. it invests. — the transmission argument (you are here)
- five kinds of headlines, zero checks — a test for the inbox
- the subsidy that proves the product — blended finance as diagnosis
- demand by statute — what a law-made buyer can and cannot do
- markets before mandates — instrument first, regulation second