Post date 17 Aug 2026 (original LinkedIn date)
Original LinkedIn version· 3,779 words
Part 5 (and likely final – can’t keep flogging a dead horse, so to speak!) of a series on discount rates, evaluation horizons, and what nature finance could actually be doing about our extreme weather challenges – drought ~ flood ~ drought ~ flood ~ drought
‘Gimmer shelter’, “released in late 1969 during the height of the Vietnam War and amid widespread civil unrest, the song’s unsettling lyrical themes and visceral vocal delivery shocked mainstream listeners and challenged radio conventions” (allegedly - even I am not quite old enough to remember!). At the time, it was considered controversial primarily for its raw, apocalyptic imagery, but were the Stones just ahead of their time?
My fourth piece in this series finished by arguing that the next stage of nature finance market design is not primarily about mobilising more capital. I genuinely believe that the capital exists, and, as we discussed, the institutional cover for deploying it has substantially opened up over the last twelve months. But what is still missing is the market infrastructure that would let the capital flow at scale, and I pointed to Dr Roger Iles‘s insurance pool idea as one worked example of the kind of architecture that needs to be built.
Let me run further with that theme now: an insurance pool underwrites something, but what is the underlying instrument that the pool is meant to be insuring? The proposal treats the underlying restoration projects as generic ecological assets (part of a broad nature intervention exercise) that need insurance wrapping. But the deeper question is what kind of contractual instrument is being wrapped in the first place? If the underlying assets are voluntary biodiversity credits or nature-based carbon credits, which is the current default path to monetisation, then the pool is essentially insuring a class of instruments whose fundamental contractibility problems are, I fear, a constraint on institutional participation.
Nature finance has a supply-side problem, as Roger rightly diagnoses. But underneath the supply-side problem there’s an instrument problem, and until the instrument problem gets addressed, the supply-side architecture may not really have anything worth building around.
What credits actually are … and aren’t
The voluntary carbon market, and its younger sibling the emerging voluntary biodiversity credit market, are built on a specific instrument structure. A project developer generates a credit representing a claim on a unit of environmental benefit, a tonne of carbon sequestered, a hectare of habitat restored, a biodiversity intactness improvement measured against a baseline. That credit is then sold, often through an intermediary, to a buyer who uses it to offset their own environmental impact or to make a sustainability claim about their operations. The value of the credit derives from its use as a claim, and the market for it is driven by the aggregate demand for claims of that kind.
I think that there are several things worth noticing about this structure. The first is that the buyer of the credit is not necessarily the party that benefits from the underlying environmental improvement. A company in Frankfurt buying credits generated by a reforestation project in Colombia is not benefiting economically from the trees. They’re benefiting from the ability to make a claim about their operations. This is not a criticism of credit buyers, most of them are acting in good faith and the claims they make are genuinely useful within corporate sustainability frameworks. Many will genuinely be part of a mitigation hierarchy process – some may not. But it means the demand side of the credit market is fundamentally driven by claim-making rather than by direct commercial benefit from the underlying ecological outcome.
How unreliable that claim-driven demand actually is has now been quantified. A study published in One Earth in July evaluated the biodiversity commitments of 180 keystone corporations around the world, the companies with the greatest influence over the biosphere, scoring 637 of their forward-looking commitments against basic verifiability criteria. 79% of the companies make some form of biodiversity pledge. Only 13% make even one commitment robust enough to be held accountable against. Barely a third of the commitments are measurable at all, fewer than a third specify a baseline, and of the measurable ones, 72% track actions taken rather than ecological outcomes. A buyer pool whose own pledges are overwhelmingly unverifiable doesn’t feel like a foundation an asset class can be built on? Just as telling is where the robust commitments cluster - typically in sectors where regulation bites, or where nature loss poses a direct operational risk to the business. Even within the voluntary pledge landscape, seriousness appears to track direct commercial exposure rather than reputational ambition.
The second is that the credit is generated ex post relative to the project. The developer establishes a baseline, undertakes the intervention, monitors the outcome, and then issues credits based on measured performance against the baseline. This creates a specific pattern of financing risk. The developer has to fund the intervention upfront against uncertain future credit generation and uncertain future credit prices. Institutional investors funding the development stage are therefore taking on some form of development risk that the credits will be generated as expected, that the market will absorb them, and that prices will hold roughly as forecast. The instrument itself does not provide much protection against any of these risks.
The third is that credits are not contractually enforceable in any meaningful sense against a specific counterparty. If a credit fails to be issued, or is issued and subsequently invalidated by a registry, the developer may not have a counterparty to sue, and the market would absorb the loss. This is very different from a bond or a loan or an insurance contract, where the party bearing the risk has a specific legal claim against a specific counterparty for a specific obligation.
None of these features is fatal to the credit market. Voluntary markets have functioned for years despite them, and there are contexts in which they are the appropriate instrument. But they do explain why the credit market has repeatedly failed to attract institutional capital at the scale that the underlying environmental need would justify. The instrument has structural properties that make it hard for institutions with fiduciary obligations to hold at scale. Every high-profile scandal about invalidated carbon credits in the last five years has been, at root, a manifestation of these structural properties.
What outcome-based contracts do differently
There is a different class of instrument that addresses these problems at the root. The core idea is pretty straightforward. Instead of a credit representing a claim on an ecological outcome that can be sold to any buyer for any purpose, you have a contract between a specific offtaker and a specific project. The offtaker is a party that receives a direct commercial benefit from the ecological outcome. The contract obligates the project to deliver a defined ecological service and obliges the buyer to pay for that service against defined performance criteria. The instrument is bilateral (or possibly multi-lateral in the presence of many offtakers) rather than essentially being a market claim.
This structure changes a number of things about the underlying economics. The offtaker’s willingness to pay is grounded in avoided cost rather than in claim value. A manufacturing site with high-value equipment exposed to catchment-scale flood risk is not paying for a biodiversity claim. They are paying for a specific reduction in the frequency and severity of flood events that would disrupt operations, damage assets, and trigger insurance claims. A hyperscaler whose data centre depends on water availability in a stressed catchment is paying for water security in the specific catchment that hosts the site. A regulated utility facing regulator-imposed obligations on catchment-scale water quality is paying for measurable improvements against those obligations. In each case, the value of the contract to the offtaker is grounded in something they were going to have to pay for anyway, either through direct expenditure on infrastructure, through insurance premiums, through regulatory penalties, or through operational disruption. The nature-based intervention is competing against those (inferior?) alternatives on the offtaker’s own commercial terms.
There is now independent evidence that this avoided cost is real, quantifiable, and visible in the language mainstream finance already speaks. Systemiq’s recent work on integrating physical climate and nature risks into credit decisions makes the point from inside the banking system. The European Central Bank has found that 72% of euro-area firms are critically dependent on ecosystem services, and that those firms account for around three quarters of euro-area corporate lending. In one of the report’s worked agricultural examples, a resilience investment lifts a borrower’s debt service coverage ratio from below 1.4 times to nearly 1.9 times, a difference any credit committee can clearly see. Yet the same report diagnoses what it calls an incentive gap: borrowers who invest in resilience typically receive no recognition for it, because credit models have no instrument through which to see the benefit. A contracted, dated payment stream for a delivered ecosystem service is exactly the kind of object a credit model can price, which is one more reason the contract, rather than the credit, is the natural unit of institutional nature finance.
The financing profile of the contract also looks different. Think of it this way - because the counterparty and the obligation are both defined, the contract can be structured in ways that credits cannot. Payment streams can be phased over the life of the contract rather than being contingent on ex post credit issuance. Performance failure can trigger defined remedies rather than being absorbed by the market. The contract itself becomes an asset with recognisable financial characteristics, which means it can be pledged, sold, or wrapped in insurance in ways that have well-established market analogues.
The measurement and verification structure is also hopefully more tractable. Outcome-based contracts require verification of specific performance criteria against specific baselines, agreed between two identified counterparties who both have ‘skin in the game’ (least favourite financial phrase ever!). Arguably this is a much more manageable measurement problem than the general question of whether a credit accurately represents a unit of environmental benefit against a hypothetical baseline for the purposes of an unspecified future buyer. The counterparties can invest in the monitoring infrastructure they need, disputes can be resolved bilaterally, and the measurement infrastructure develops in response to specific contractual requirements rather than in the abstract.
The Resilience Purchase Agreement
Some of the work Rebalance Earth has been developing over the last year or two has been a specific worked example of this class of instrument, which we’ve given the working name of the Resilience Purchase Agreement or RPA. Let me describe it briefly, not because the specific product is the point, but because a concrete example should help ground what would otherwise be an abstract argument about instrument design.
An RPA is a contract between a nature-based restoration project and an offtaker who receives a direct commercial benefit from the resulting ecosystem service, structured around three linked layers. A Framework Agreement establishes the general contractual architecture, the parties, the definitions, the enforcement mechanisms, and the general obligations, in a way modelled on the ISDA Master Agreement structure that underpins the derivatives market. A Supplement specifies the particular ecosystem service being contracted for, the delivery obligations, the verification framework, and the payment schedule for a specific transaction under the Framework. Appendices contain the technical detail on measurement, baselines, and site-specific parameters. All fairly straight forward and hopefully eminently replicable without great hassle!
The three-layer structure is doing important work - it separates the general contractual infrastructure, which is amortised across many transactions and can be standardised, from the transaction-specific parameters, which necessarily vary from deal to deal. This is exactly the pattern that allowed the derivatives market to scale from bespoke bilateral arrangements to a standardised institutional asset class. The specific counterparty categories the RPA framework has been developed for tell you something about the underlying commercial logic: manufacturing sites with high-value equipment exposed to catchment-scale flood risk, hyperscalers whose data centres depend on water availability in stressed catchments, regulated utilities facing regulator-imposed obligations on catchment resilience, insurers looking to hedge specific catchment-scale physical risks in their underwriting portfolios, and rail and grid operators whose infrastructure is exposed to nature-driven disruption. In each case, the offtaker has a specific commercial exposure that nature-based intervention can materially reduce, and the RPA structure allows that avoided cost to be contracted for directly.
The point of describing this is not to argue that the RPA framework as we’ve developed it is the only or even the best worked example of what outcome-based nature contracts should look like. I’m certain that other frameworks are being developed in parallel by other institutions, and it would be great if the market could coalesce around the best. The point is to make concrete what the class of instrument looks like when it’s actually designed properly, so that the abstract argument about credits versus contracts has a specific worked example to point at.
Why this matters
Roger Iles’s insurance pool proposal needs a class of underlying contracts to insure. If the underlying contracts are voluntary credits, the pool is trying to underwrite an instrument whose contractibility problems it cannot fully fix. If the underlying contracts are outcome-based agreements with identified offtakers, the pool is underwriting delivery risk against defined counterparty obligations, which is a much more tractable problem. To some extent, the choice of instrument determines what the insurance architecture can actually do, and actually the protection gap makes the point sharper still. Only around a quarter of natural catastrophe losses in the European Economic Area are currently insured, and in agriculture the insured share is lower again. Where insurance retreats, the residual risk lands on lenders, asset owners, and operators, which is exactly the exposed constituency whose avoided costs an outcome-based contract gives the pool to organise.
The Gollier-Drupp framework says ecosystem service values should be projected as rising in real terms over time. In the credit market, this rising value has no natural mechanism for reaching the developer of the underlying project, because the credit price is set by aggregate market demand for claims rather than by the marginal value of the ecosystem service to a specific beneficiary. In the outcome-based contract structure, the rising value can be built into the contract itself through indexation clauses (inflation-linkage), floor pricing mechanisms, or renegotiation triggers. The instrument structure determines whether the theoretical scarcity premium can actually be captured commercially.
The IRR truncation problem is also affected by the instrument choice. Credit-based projects face short-cycle IRR appraisal because the credits are being issued and sold in relatively short cycles even if the underlying ecological outcomes are long-duration over many decades in to the future. Contract-based projects can have payment streams and performance windows structured to match the natural time horizon of the ecosystem service being delivered, which brings the appraisal window into alignment with the underlying ecology. Now the instrument determines whether the appraisal framework can even see the long tail.
The Cushon fiduciary opinion also lands differently depending on the instrument. Trustees who are asked to consider the long-term systemic context beneficiaries retire into can more easily defend contracting for ecosystem services with defined counterparties than they can defend holding credits whose commercial value depends on the persistence of voluntary market demand. The choice of instrument shapes what the broader fiduciary interpretation can actually support in portfolio terms.
In short, the class of instrument that nature finance defaults to determines what the surrounding architecture, insurance, measurement, appraisal, fiduciary interpretation, can actually be built to support. Credits have to some extent limited what all of these adjacent developments could accomplish, because the underlying instrument has structural properties that don’t necessarily play well with institutional-scale finance. Outcome-based contracts have very different properties that open up architectural possibilities that the credit-based framing has been foreclosing.
The intellectual lineage
This line of thinking is not original to Rebalance Earth or to the specific RPA framework we’ve been describing. It draws on a body of work that has been developing over the last several months and years, so let me acknowledge the sources properly.
Sebastian Mynott‘s argument about the biophysical geometry of ecosystem services is one of the intellectual foundations. His observation that ecosystem services are inherently local, spatially bounded, and tied to specific beneficiaries points directly at why bilateral contracts are the natural instrument structure and why market-generic credits will always struggle to capture what’s actually happening ecologically. Prof Nicola Ranger (and team)’s work on catchment-scale resilience investment is another. The Sophus zu Ermgassen work on nature-related contracting frameworks provides much of the design vocabulary. The ISDA Master Agreement architecture, which underpins the entire derivatives market, is the analogue for how contractual standardisation allows a bespoke market to scale.
RPAs are attempts to translate these intellectual foundations into working commercial products. The point is not that any single framework has solved the problem, rather the point is that the class of instrument being pursued through this line of thinking is structurally different from the credit-based default, and it addresses a set of problems that credit-based approaches have thus far been unable to solve.
Where does this leave us
The four earlier pieces in this series argued that nature finance has been struggling to achieve take off velocity because of a specific set of interlocking constraints: discount rate methodology that undervalues long-tail nature assets, evaluation horizons that hide most of the appreciating value, fiduciary interpretations that limit what trustees can do, and supply-side architecture that doesn’t yet exist at scale. I’ve been writing that the intellectual and institutional infrastructure to fix these constraints has been slowly appearing over the last few years, and that the response from the asset management industry has been slower than the underlying case warrants. The Green Book Discount Rate Review published in June, which recommends cutting the headline UK social discount rate from 3.5 per cent to 3.0 per cent, is only the most recent arrival, and a reminder that the methodological ground is moving in exactly the direction my first piece in this series argued it should.
This final article adds one more constraint that hasn’t quite been properly grounded in the series so far. The default class of instrument that nature finance has been built around, voluntary credits, has structural properties that make it very difficult to build the surrounding architecture the earlier pieces have been arguing for. Insurance wrappers, appraisal reforms, fiduciary reinterpretations, and universal owner frameworks all work more naturally around contract-based instruments than they do around credit-based ones. The instrument choice is not a technical detail; instead, it’s the foundation the rest of the architecture has to sit on.
The class of outcome-based contracts we’ve been describing (the RPA framework is one worked example) may not be the finished answer, but I think it’s the beginning of an answer, and it will need to be developed further by many hands working in parallel. It does however point in the direction the market design has to go if the reforms the earlier pieces have been advocating are actually going to translate into deployed capital at scale.
I no longer believe that the credit-based default is on its own sustainable as the primary vehicle for institutional-scale nature investment. Something else is needed, and the shape of what that something else has to look like is becoming clearer. The next few years of market design will determine whether the sector actually builds it, or whether it continues trying to force credit-based instruments to bear a weight I don’t think they were ever designed to hold.
References
Fundamental
Mynott, F.J. (2026). “The biophysical geometry of ecosystem services.” LinkedIn article.
Ranger, N., Alvarez, J., Freeman, A., Harwood, T., Obersteiner, M., Paulus, E. and Sabuco, J. (2023). The Green Scorpion: the Macro-Criticality of Nature for Finance. Foundations for scenario-based analysis of complex and cascading physical nature-related risks. Environmental Change Institute, University of Oxford; co-published as an Oxford-NGFS Occasional Paper, December 2023. https://www.eci.ox.ac.uk/sites/default/files/2023-12/INCAF-MacroCriticality_of_Nature-December2023.pdf
University of Oxford Environmental Change Institute and Global Center on Adaptation (2023). Financing Nature-Based Solutions for Adaptation at Scale. https://www.eci.ox.ac.uk/sites/default/files/2023-12/Financing_NbS_for_Adaptation-GCAOxford2023-finalv2.pdf
zu Ermgassen, S.O.S.E., Marsh, S., Ryland, K., Church, E., Marsh, R. and Bull, J.W. (2021). “Exploring the ecological outcomes of mandatory biodiversity net gain using evidence from early-adopter jurisdictions in England.” Conservation Letters, 14(6), e12820. https://doi.org/10.1111/conl.12820
On the credits versus contracts distinction and voluntary market limitations
West, T.A.P., Wunder, S., Sills, E.O., et al. (2023). “Action needed to make carbon offsets from forest conservation work for climate change mitigation.” Science, 381(6660), 873-877. https://doi.org/10.1126/science.ade3535
Guizar-Coutiño, A., Jones, J.P.G., Balmford, A., Carmenta, R. and Coomes, D.A. (2022). “A global evaluation of the effectiveness of voluntary REDD+ projects at reducing deforestation and degradation in the moist tropics.” Conservation Biology, 36(6), e13970. https://doi.org/10.1111/cobi.13970
Trouwloon, D., Streck, C., Chagas, T. and Martinus, G. (2023). “Understanding the use of carbon credits by companies: a review of the defining elements of corporate climate claims.” Global Challenges, 7(4), 2200158. https://doi.org/10.1002/gch2.202200158
Hawkins, I., Bromwich, T., Jouffray, J.-B., White, T.B., Bull, J.W., Chang, F., Milner-Gulland, E.J. and zu Ermgassen, S.O.S.E. (2026). “Evaluating the biodiversity commitments of Earth’s keystone corporations.” One Earth, 9, 101747. https://doi.org/10.1016/j.oneear.2026.101747
Adjacent instrument frameworks and case examples
Efficiency Valuation Organization (2022). International Performance Measurement and Verification Protocol: Core Concepts. EVO, Washington, DC. Verification protocol referenced in RPA design work. https://evo-world.org/en/products-services-mainmenu-en/protocols/ipmvp
International Swaps and Derivatives Association (2022, version 3.0 2025). Verified Carbon Credit Transaction Definitions. https://www.isda.org/book/2022-isda-verified-carbon-credit-transaction-definitions/
Systemiq (2026). Integrating Physical Climate and Nature Risks into Credit Decisions. Nature on the Balance Sheet initiative, June 2026.
https://www.systemiq.earth/
Demand-side and supply-side framework from earlier pieces
Drupp, M.A., Hänsel, M.C., Fenichel, E.P., Freeman, M., Gollier, C., Groom, B., Heal, G.M., Howard, P.H., Millner, A., Moore, F.C., Nesje, F., Quaas, M.F., Smulders, S., Sterner, T., Traeger, C. and Venmans, F. (2024). “Accounting for the increasing benefits from scarce ecosystems.” Science, 383(6687), 1062-1064. https://doi.org/10.1126/science.adk2086
Eversheds Sutherland for NatWest Cushon (2025). Legal opinion on pension trustee fiduciary duty and members’ future living standards. Unveiled 5 March 2025. Reported in Pensions Age: https://www.pensionsage.com/pa/legal-opinion-provides-further-clarity-on-fiduciary-duty-rules.php
HM Treasury (2026). The Green Book: Central Government Guidance on Appraisal and Evaluation. February 2026. https://www.gov.uk/government/publications/the-green-book-appraisal-and-evaluation-in-central-government
Groom, B. and Freeman, M. (2026). Green Book Discount Rate Review: Findings and Recommendations. HM Treasury, June 2026. https://www.gov.uk/government/publications/green-book-discount-rate-review-2026
Iles, R. (2026). “Making Nature Restoration Investable: Why a corporate-backed insurance pool could unlock private capital for restoration.” LinkedIn article
Systemic framing
Jones, A., Bedenham, G., Goldman, M., Ranchin, A., Spencer, N. and Trim, I. (2026). Planetary Solvency: Tipping into the wild unknown. Global nature risk management. Institute and Faculty of Actuaries and Anglia Ruskin University, April 2026. https://actuaries.org.uk/planetary-solvency-tipping-into-the-wild-unknown
Dasgupta, P. (2021). The Economics of Biodiversity: The Dasgupta Review. HM Treasury, UK Government. https://www.gov.uk/government/publications/final-report-the-economics-of-biodiversity-the-dasgupta-review