Post date 8 May 2026 (original LinkedIn date)
Original LinkedIn version· 2,694 words
“Take the long way home“ by Supertramp, released in 1979 on the album Breakfast in America
Sir David Attenborough turned a hundred this week. He has been broadcasting on the natural world for over seventy years, which is roughly four times the standard evaluation horizon for an institutional investment in nature. It is that mismatch that I want to focus on today. Most investment committees, when they appraise a nature-related project, do so over a window of fifteen or twenty years. They treat that horizon as a modelling convenience, a risk-acceptable length of time for investing, but it isn’t, it’s neither of these things really. For nature-related assets, it is a significant assumption in our investment appraisal, and Attenborough’s career is, by accident, a useful demonstration of why.
The natural world often does not make sense over such short horizons. Forests mature over centuries, soil regenerates over decades, peat accumulates over millennia, and the consequences of human action on ecosystems take a generation or longer to become visible. Attenborough has been showing us this for seventy years. The fact that our financial frameworks still cannot see what he has been pointing at is not a small problem. Let’s take the long way home.
Earlier this week I argued that the discount rate debate in nature finance as a debate has been convincingly won. The theoretical questions are settled, even if its not yet widely implemented. Gollier wrote the proper formalism more than a decade ago; Drupp and co-authors translated it into a policy-ready rule in 2024. Governments are slowly adopting it; the remaining contention is operationalisation rather than intellectual. I finished by suggesting that there’s a sharper question hiding underneath the discount rate one, and that this question, once you actually look at it, is where I believe the real distortion in capital allocation toward nature lives. This is that piece.
I want to make a fairly strong claim. Investment committees make assumptions of convenience as part of their appraisal of nature-related asses - one is the discount rate. Another is the evaluation horizon. The fact that most projects get assessed on fifteen or twenty-year IRR is a modelling convenience, a default, a reflection of fund life or asset depreciation. In fact it is a substantive position about how the future works, and for nature-related assets it is almost always wrong.
The reason it goes unexamined is that, unlike the discount rate, it doesn’t appear as a parameter to argue about. It appears as the boundary of the analysis itself. You don’t see it on the page of the investment memo. You see only what is left, because the rest has been filtered out.
What a finite-window IRR is actually doing
Let me try to properly explain this. When you evaluate a project on twenty-year IRR, three things happen at once.
The first is truncation. Everything beyond year twenty either gets a terminal value (a massively discounted exit of future cash flows), typically driven by a Gordon growth model with some small constant rate. For a data centre, a fleet of trucks, a piece of industrial equipment, this is fine. The asset depreciates, the technology becomes obsolete, the residual is small relative to what’s already in the explicit period. The truncation costs you almost nothing because there isn’t much value left to truncate.
For a peatland, a regenerated wetland, a reforestation project, an oyster reef, a programme of soil restoration, the truncation is important, and not just in the way most people assume. The intuition is that you’re losing the long tail of cash flows. That’s true, but it’s not the main thing., which is rather that the cash flows you’re losing are appreciating in real terms, and the further out you go the steeper the appreciation gets. You’re not just deleting cash flows; you’re deleting the part of the value that distinguishes nature from a substitutable industrial asset.
This connects directly to the argument from earlier this week. The Gollier-Drupp framework says that the willingness-to-pay for ecosystem services should be projected as rising over time at a rate of ξ × (g_C − g_E). Apply that adjustment to a project but cut the model at year twenty and you’ve done a strange thing. You’ve acknowledged that nature appreciates, then you’ve thrown away most of the appreciation by truncating the part of the time series where the appreciation actually compounds. The discount rate fix and the truncation fix are linked. Doing the first without the second is just half a job.
The second thing that happens is that the value trajectory gets collapsed into a single number. IRR is, mathematically, the discount rate that sets NPV to zero. It’s a summary statistic of the cash flow shape. For roughly stationary cash flows it’s a serviceable summary. For non-stationary ones, particularly ones where the underlying value is rising, it actively misleads. Two projects with identical IRRs can have entirely different value trajectories, and the metric meant to compare them suppresses precisely the dimension on which they differ.
The third thing (and this is where the real damage happens) is that the project gets compared to alternatives evaluated on the same window with the same metric. Investment committees don’t appraise projects in isolation. They appraise them against a queue. A nature-positive project showing 6% IRR over twenty years and a conventional infrastructure project showing 9% IRR over twenty years aren’t being compared on equal terms. They’re being compared on terms that systematically suppress the part of the nature-positive project’s value that exists in years twenty-one through, say, fifty. Extend the window, apply the scarcity-adjusted shadow prices to the nature-related cash flows in both projects (because the conventional project almost certainly has nature-related cash flows hidden in its supply chain dependencies), and the rankings can flip. Not because the underlying economics changed., but because the framework was suppressing the dimension that mattered.
This isn’t a hypothetical point. We’re describing the actual mechanism by which institutional capital has under-flowed to nature for decades despite a steady accumulation of evidence about ecosystem service value. The evidence didn’t fail to register; its just that the framework filtered it out before it reached the point where decisions get made.
Why this matters perhaps even more than the discount rate
You can have any view you like on whether the appropriate discount rate for nature is 1.5% or 4%. Within a twenty-year window, the difference is real but bounded. A cash flow at year eighteen discounted at 4% versus 1.5% changes by a factor of about 1.6. A cash flow at year seventy five, by contrast, changes by a factor of about nine. The longer-dated cash flows are where the discount rate argument actually bites. Those are exactly the cash flows the 15-year IRR window has already deleted. This means that the discount rate debate is about a future the IC has already decided not to look at.
This is part of why the discount rate debate, despite its intellectual depth, has produced relatively limited change in practice. Investment committees can absorb a sensitivity analysis on the rate without their world-view shifting. Asking them to reconsider the evaluation horizon is harder, because the horizon isn’t a parameter inside the model, it’s the framing within which the model exists. Changing it forces a different conversation about what’s actually being measured.
The duration mismatch nobody is talking about
The strongest version of this argument is fiduciary. I think it hasn’t been properly articulated yet, and I expect it will over the next decade. Pension trustees, life insurers, sovereign wealth funds, university endowments, all of them hold liabilities that extend thirty, forty, fifty years out, sometimes longer. The fiduciary obligation runs the length of the liabilities. The whole architecture of liability-driven investing exists precisely because matching duration to obligation matters.
But the same institutions routinely evaluate the assets meant to discharge those liabilities on fifteen or twenty-year IRR windows. This is, I think,, peculiar. The institution declares a fifty-year obligation, then assesses candidate assets on a framework that systematically truncates value beyond year twenty. For most assets, equities, bonds, real estate, the mismatch is muted, because the explicit-period IRR is a reasonable proxy for the long-run return profile. For nature-related assets it isn’t, because the truncation is selectively deleting the part of the return profile that the long-duration investor most needs.
What this amounts to is a duration mismatch hidden inside the appraisal methodology rather than disclosed on the asset-liability statement. It doesn’t appear in any ALM report. It doesn’t show up in any duration calculation. It’s invisible because it’s structural - it’s a feature of how candidate assets get filtered before they ever reach the asset-liability conversation.
I haven’t seen this articulated as a fiduciary case in any jurisdiction yet, but I’d be surprised if it didn’t become one. I feel the argument is fairly straightforward. Trustees are obliged to consider the long-term interests of beneficiaries. The standard appraisal framework systematically underweights long-term interests for the specific class of assets best matched to long-term obligations. Therefore, the standard framework is inconsistent with the fiduciary duty for long-horizon investors.
ClientEarth-style litigation has been built on weaker foundations than this. Somewhere, probably the UK or the Netherlands, a pension trustee will get asked why they used a twenty-year IRR to assess a nature-related asset on behalf of a fifty-year liability. The answer “because that’s what we always do” is unlikely to survive that question.
What changes if you take this seriously
Three things can be changed, sadly none of them comfortable with current practice.
The evaluation horizon stops being a default and becomes a defended choice. A twenty-year window for a nature-positive asset isn’t neutral; it’s a substantive position that everything happening after year twenty is either zero or negligible. For a peatland that sequesters carbon for centuries, or a soil restoration project whose ecological time constant is measured in decades, or a reforestation programme whose mature value is realised in years thirty through eighty, the twenty-year window is making a strong claim about what doesn’t matter. That claim should be justified, not assumed. Investment memos should defend the truncation point, not inherit it from a template.
Terminal values need to do real work, in the opposite direction from current practice. Standard terminal value calculations fade to mature growth ie they assume the asset’s growth rate slows and stabilises in the post-explicit period. For nature-related assets under the Gollier-Drupp logic, the terminal-period scarcity premium is larger than the explicit-period premium, because the underlying ecosystem service has become scarcer. The terminal value should arguably grow faster than the explicit-period cash flows, not slower. Done properly, it can double or triple the valuation of long-lived nature assets.
IRR may simply be the wrong primary metric. IRR has known mathematical pathologies that have been written about for sixty years - ambiguity with sign-changing cash flows, the implicit reinvestment-at-IRR assumption, problems with comparability across projects of different durations. For nature-related assets, where the value of holding through time is part of the return, you need a metric that surfaces the rising-value path rather than collapsing it. How about a modified IRR with a scarcity-adjusted reinvestment rate?Or asymmetric return measures that distinguish nature-appreciating from nature-depleting holdings? I don’t think that either of these are particularly exotic, they’re just unfamiliar in current investment committee practice.
A practical first move
I don’t think this needs a wholesale framework rewrite. The first move is much smaller and entirely tractable, and I’d suggest it as something any investment committee could do tomorrow.
For projects with material nature exposure, and that includes nature-positive assets and nature-dependent assets, the second category being far larger than most committees realise, run the DCF twice. Once on the standard fifteen or twenty-year window with the conventional assumptions you’d use for any project. Once on a fifty-year window with scarcity-adjusted shadow prices applied to the nature-related cash flows, using something close to the Drupp default of WTP rising at ξ × (g_C − g_E) with ξ at one and g_E calibrated to whichever ecosystem service trajectory is most defensible for the asset in question.
The gap between the two valuations is the implicit position the committee is taking on whether nature is substitutable. Today that position is invisible, it’s buried in the methodology. Making it explicit, even without resolving which valuation is ‘correct’, changes the conversation. Some committees will look at the gap and conclude the standard window is fine for their purposes. Others will discover they’ve been systematically rejecting the assets best matched to their long-term obligations. Both are useful outcomes, and either way, the assumption gets surfaced rather than hidden.
This is the change with the highest value relative to the institutional disruption involved. It doesn’t require new theory, it doesn’t require regulatory change, and it doesn’t even require committees to agree which valuation is right. It just requires them to look at both numbers.
Where this leaves us
The discount rate debate is important. Gollier won the theoretical argument. Drupp and colleagues wrote the implementation guide. Governments are slowly adopting it. But we can go furhter. The truncation debate hasn’t really started. The fifteen-to-twenty-year IRR window is the deepest unexamined assumption in long-horizon investment practice, it has arguably the largest distortive effect on capital flows to nature, and the fiduciary case against it is,strong. Yet you can search the academic literature on nature finance and find almost nothing on it - the rate gets all the attention and the window gets none.
This is where the next decade of consequential argument is going to happen, and it won’t happen on conference panels or in peer-reviewed journals. It will happen in investment committee minutes, in pension trustee training, in the second-order valuations that allocators run alongside their primary models, and eventually, I think, possibly in court. The good news is that the technical fix is small. The hard part is that the fix forces investment committees to look at something they have been arranging not to see! The truncation is the real problem. The discount rate argument was its rehearsal.
References
Foundational sources for the truncation argument
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
Gollier, C. (2010). “Ecological discounting.” Journal of Economic Theory
On IRR’s limitations and alternative metrics
Hazen, G.B. (2003). “A new perspective on multiple internal rates of return.” The Engineering Economist
Magni, C.A. (2010). “Average internal rate of return and investment decisions: a new perspective.” The Engineering Economist
Phalippou, L. (2008). “The hazards of using IRR to measure performance: the case of private equity.” Journal of Performance Measurement
On tipping points and the limits of smooth-substitution frameworks
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.
Trust, S., Saye, L., Bettis, O., Bedenham, G., Hampshire, O., Lenton, T.M., and Abrams, J.F. (2025). Planetary Solvency — finding our balance with nature. Institute and Faculty of Actuaries.
Marsden, L., Ryan-Collins, J., Lenton, T.M. and Abrams, J.F. (2024). Ecosystem tipping points: Understanding the risks to the economy and the financial system. UCL Institute for Innovation and Public Purpose, Policy Report 2024/03.
On nature-related financial risk and fiduciary duty
Ranger, N., Oliver, T., Alvarez, J., et al. (2024). Assessing the materiality of nature-related financial risks for the UK. Green Finance Institute.
Kedward, K., Ryan-Collins, J., and Chenet, H. (2023). “Biodiversity loss and climate change interactions: financial stability implications for central banks and financial supervisors.” Climate Policy
Taskforce on Nature-related Financial Disclosures (2023). Recommendations of the Taskforce on Nature-related Financial Disclosures.
On long-horizon investor frameworks
Hawley, J.P. and Williams, A.T. (2007). “Universal owners: challenges and opportunities.” Corporate Governance: An International Review
Mercer (2015). Investing in a Time of Climate Change. Mercer Investment Consulting.
On measurement and the connection to operationalising scarcity-adjusted projections
Goldman Sachs and MIT-IBM Watson AI Lab (2026). AI for Biodiversity Measurement: Advancing Nature Finance.