An overview of global biodiversity finance

Annex A. Detailed methodology and data sources

This annex describes the data sources, inclusion rules, attribution methods and calculations used to estimate biodiversity finance. The aggregate estimate combines four broad components:

• domestic public biodiversity expenditure;

• international public biodiversity finance;

• return-seeking domestic and international private investment; and

• non-return-seeking domestic and international private finance, including philanthropy and expenditure by selected biodiversity-focused non-governmental organisations.

The estimate measures financial resources intended to contribute to biodiversity conservation, sustainable use or restoration. It does not assess the effectiveness, additionality or biodiversity outcomes of the financed activities, and it does not deduct financial flows that may be harmful to biodiversity.

Several related indicators are excluded from the total and presented separately. These include bond issuance, assets under management, finance mobilised by official development interventions, corporate environmental-protection expenditure and transactions in environmental markets. They are excluded because they measure different financial concepts, may represent different stages of the same financing chain, or may overlap with finance flows included elsewhere in the aggregate estimate.

Reference period and annualisation

The analysis uses observations from 2023-25. The period begins immediately after the adoption of the Kunming–Montreal Global Biodiversity Framework in December 2022 and balances recency with data availability. Restricting the assessment to 2025 would substantially reduce coverage because many official and private datasets are published with a reporting lag.

For each source and reporting unit, the arithmetic mean of the annual observations available within 2023–25 was calculated:

• where data for 2023, 2024 and 2025 were available, a three-year average was used;

• where two annual observations were available, their average was used; and

• where only one observation was available, that value was retained as the best available indication of the annual level during the reference period.

Missing annual observations were not imputed. The resulting aggregate should therefore be interpreted as the estimated average annual level during 2023-25 based on available observations, rather than as a balanced panel estimate or an estimate for a specific calendar year.

A similar approach was applied to trend analysis. Average annual values for 2018-20 were used as the baseline and compared with average annual values for the most recent period available. Where data were not available for every year within a period, averages were calculated from the available observations. Trend comparisons were restricted to countries or financial flows with sufficiently consistent coverage and methodology across both periods.

Currency conversion and price basis

All values are expressed in constant 2024 United States dollars. Values reported in national currencies were converted using annual exchange rates for the corresponding observation year and adjusted to 2024 prices using World Bank exchange-rate and price-index data. Where underlying estimates had already been converted to USD, the reported USD values were adjusted to 2024 prices using the relevant price index as a proxy.

Interpretation of estimate ranges

The ranges presented in the report are methodological scenarios, not statistical confidence intervals. For domestic public expenditure, the range reflects the inclusion or exclusion of expenditure reported by the EU to the CBD.

Domestic public expenditure

The estimate of public domestic biodiversity expenditure draws on three sources: (i) National Reports submitted to the Convention on Biological Diversity (CBD); (ii) Government Finance Statistics classified according to the Classification of the Functions of Government (GFS-COFOG); and (iii) Environmental Protection Expenditure Accounts (EPEA), classified according to the Classification of Environmental Purposes (CEP). Further information on each source and the methodology used is provided below. A comparison of data for countries covered by at least two of these data sources shows no consistent bias between the datasets, with neither source systematically reporting higher or lower values than the other (see Figure A A.1). This is despite the fact that data reported to the CBD are supposed to cover direct and indirect biodiversity finance, whereas GFS-COFOG and EPEA-CEP track expenditure with biodiversity as a primary purpose.

Only data observations falling within 2023-25 were included in the estimate. No expenditure was imputed for countries without data.

Figure A A.1. Comparison of figures across three datasets covering domestic public biodiversity expenditure

Note: The figure includes countries with 2023 data available from at least two of the three data sources used in the analysis: 7th National Reports (NR), COFOG 05.4, and EPEA CEP 0502. Estimates reported in National Reports are indexed to 100. The Philippines, Türkiye, Thailand, and Norway were excluded because the estimates were more significantly different, making them outliers and skewing the figure. For the Philippines, Türkiye and Norway the National Report data were lower, while for Thailand it was higher than COFOG.

Source: Authors.

Direct and indirect biodiversity expenditure

CBD national reports may include direct and indirect biodiversity-related expenditure. Where countries reported separate direct and indirect values, both were included. Where a combined total was reported, the full reported value was retained.

Country methodologies differ. Some countries apply attribution coefficients to programmes for which biodiversity is an indirect objective; others either report the full value of such programmes or do not specify. CBD-derived values are therefore not comparable across all countries.

COFOG Group 05.4 and CEP Group 0502 classify expenditure according to its primary purpose. For this analysis, expenditure from these sources is similar to what the CBD refers to as direct biodiversity expenditure. These categories generally exclude expenditure recorded under other functions, such as agriculture, forestry, fisheries, water management or pollution abatement, even where it may produce biodiversity benefits (albeit not as a primary purpose).

Combining CBD, GFS-COFOG, EPEA-CEP observations

Two source-selection approaches were applied. Under the aggregate approach, CBD national-report data are prioritised. Where no usable CBD observation was available for a country during 2023-25, the gap was filled using GFS-COFOG data. EPEA-CEP data were used where neither a usable CBD nor COFOG observation was available. This order gives priority to biodiversity finance reported by countries to the CBD while using GFS-COFOG as the more widely available government-expenditure classification and EPEA-CEP as an additional source for mainly EU and OECD countries. When combined the estimate comprises data from 7^th National Reports from 61 countries and the EU; GFS-COFOG data from 40 countries; and EPEA-CEP data from 2 countries.

Under the maximum-value approach, the highest available observation for each country across CBD, COFOG and EPEA was used. This sensitivity analysis addresses cases in which CBD national reports explicitly cover only selected ministries, programmes or projects. It may reduce the risk of underestimation. However, as it presented as an alternative estimate rather than the preferred approach because countries did not choose to report these data in their 7th National Reports.

EU institutional expenditure is added separately in the upper scenario. As available data do not allow all transfers between EU institutions and Member States to be traced, some EU expenditure may also be included in national figures. The upper scenario may therefore contain some double counting.

Public domestic biodiversity expenditure

National reports to the CBD

The CBD reporting framework allows Parties to report expenditure together with the year, currency, unit and methodological information. The 142 Seventh National Reports to the CBD available by mid-September 2026 were analysed and relevant data extracted. Country-reported data were retained even where methodological information was incomplete, provided the currency and unit were provided, and it was clearly recorded as biodiversity expenditure rather than broader environmental expenditure. Data reported as a cumulative total over a time period beyond 2023-25 was excluded.

Government Finance Statistics and COFOG (collected by Eurostat, OECD and IMF)

COFOG is an international classification of government expenditure by purpose. It supports comparison of government expenditure across countries and over time and is used within both the System of National Accounts and the System of Environmental-Economic Accounting (UNSD, 2000[13]).

COFOG is structured around 10 broad policy objectives or “divisions” (e.g. general public services, defence, environmental protection) which is complemented with “functional” groups or sub-sectors of expenditures that contribute towards the achievement of the policy objectives. Under Division 05 on Environmental Protection there are six groups, of which Group 05.4 is biodiversity-specific (see Table A A.1). The breakdown of Environmental Protection is based upon the Classification of Environmental Protection Activities (CEPA) (UNSD, 2000[13]). Group 05.4 on Protection of biodiversity and landscape covers (Eurostat, 2019, p. 213[17]):

[A]ctivities relating to the protection of fauna and flora species (including the reintroduction of extinct species and the recovery of species menaced by extinction), the protection of habitats (including the management of natural parks and reserves) and the protection of landscapes for their aesthetic values (including the reshaping of damaged landscapes for the purpose of strengthening their aesthetic value and the rehabilitation of abandoned mines and quarry sites).

Table A A.1. First and second level COFOG Division No. 5

First-level Division: the broad objective	Second-level Groups: the means by which first-level objectives are achieved.
05 Environmental Protection	05.1 Waste management; 05.2 Water waste management; 05.3 Pollution abatement; 05.4 Protection of biodiversity and landscape; 05.5 Research and development environmental protection 05.6 Environmental protection, not elsewhere classified (n.e.c.)

Note: n.e.c. is not elsewhere classified.

Source: (EUROSTAT, 2011[18]; UNSD, 2000[13]).

Second-level COFOG data were compiled from internationally disseminated Government Finance Statistics collected by the IMF, OECD and Eurostat. Only expenditure classified under Group 05.4 was included. Expenditure classified under other COFOG functions was excluded, even where it may indirectly benefit biodiversity.

Preference was given to the general government sector, which provides consolidated expenditure across central, state and local government where these levels are reported. Where general government data were unavailable, central government or budgetary central government data were used instead. Separate expenditure series for different sectors or levels of government were not added together, as the absence of a consolidated general government series may indicate that the underlying data have not been consolidated across government entities. Summing these series could therefore introduce overlaps and create a risk of double counting.

Regarding data availability, second-level COFOG data on Protection of biodiversity and landscape (05.4) expenditure for 2023-25 are available for 72 countries (see Annex B). The consolidated data are publicly available on the IMF website, and are collected by Eurostat, OECD and IMF.1

EPEA-CEP (collected by Eurostat and OECD)

Environmental Protection Expenditure Accounts (EPEA) measure expenditure by government, corporations, households and non-profit institutions on activities aimed at preventing, reducing or eliminating environmental degradation. They are aligned with national accounts concepts and organise environmental expenditure using a functional classification.

The Classification of Environmental Purposes (CEP), developed under the leadership of Eurostat and adopted in March 2024 by the UN Statistical Commission, integrates and replaces two previous functional classifications used in the System of Environmental-Economic Accounting - Central Framework (SEEA-CF), the former Classification of Environmental Protection Activities (CEPA) and Classification of Resource Management Activities (CReMA). CEP is a functional classification used in the SEEA-CF, which is aligned with the System of National Accounts (SNA). CEP Environmental activities in the CEP are defined as “economic activities, products, expenditure and other transactions whose purpose is environmental protection or more efficient management of natural resources” (UNSTATS, 2024[19]).

The biodiversity category used is CEP Level II Group 0502, “Protection of biodiversity and landscape, which covers “activities, expenditures and products aimed at protecting and replenishment of wild fauna and flora, safeguarding and restoring their habitats, ecosystems of which they are part, natural or semi-natural terrestrial, marine and other aquatic ecosystems” (UNSTATS, 2024, pp. 53-54[19]).2 Table A A.2 provides an overview of the environmental purpose Division 05 “soil, surface and groundwater, biodiversity and forest”.

Data for at least one year in the period 2023-2025 are available for 21 countries. See Annex C for further detail on definitions, as well as additional CEP groups and classes that may be considered relevant for biodiversity.

Table A A.2. Classification of Environmental Purpose

Level I Division	Level II Group	Level III Class
05	Soil, surface and groundwater, biodiversity and forest
	0501	Protection of soil, surface and groundwater
		050101	Prevention of pollutant infiltration into soil and water bodies
		050102	Cleaning up of soil and water bodies
		050103	Protection from erosion and other physical degradation of soil and water
		050104	Prevention and remediation of soil and groundwater salinity
		050105	Monitoring and measurement for soil, surface and groundwater protection
		050199	Others for soil, surface and groundwater protection, n.e.c.
	0502	Protection of biodiversity and landscape
		050201	Protection and rehabilitation of species and habitat
		050202	Protection of natural and semi-natural landscapes
		050203	Monitoring and measurement for protection of biodiversity and landscape
		050299	Others for protection of biodiversity and landscape, n.e.c.
	0503	Management of forest resources
		050301	Reforestation, afforestation and forest-related land management
		050302	Protection against forest fires
		050303	Monitoring and measurement of forest resources
		050399	Others for management of forest resources, n.e.c.

Source: (UNSTATS, 2024[19]).

The internationally comparable EPEA aggregate combines government and non-profit institutions serving households. General-government expenditure could not be isolated consistently across the available datasets, so the estimate may include some Non-Profit Institutions Serving Households (NPISH) expenditure. Available country evidence suggests that this component is small; in Portugal, where it was separately identifiable by comparing EPEA data with the data they reported to the CBD, NPISH represented less than 1% of the combined value.

International public biodiversity finance

Data sources for international public finance: The OECD CRS and the TOSSD

Data on biodiversity-related development finance is available through the OECD Creditor Reporting System (CRS) and the Total Official Support for Sustainable Development (TOSSD) (statistical framework managed by the International Forum on TOSSD, independent entity hosted by the OECD). The CRS captures official development assistance (ODA), other official flows (OOF), and private sector instruments (PSI).3 International public finance data covers bilateral providers from the OECD Development Assistance Committee (DAC), multilateral development banks and other multilateral institutions. TOSSD includes cross-border flows to developing countries and regional and global expenditures in support of international public goods. It covers some additional bilateral and multilateral providers to the CRS, mainly South-South co-operation providers. Only TOSSD financial flows not reported to CRS are included in the estimate to avoid risks of double counting with data in the CRS.

Identifying biodiversity finance and application of coefficients

For activities reported through the CRS, biodiversity relevance is identified primarily using the biodiversity Rio marker. Activities marked principal are those for which biodiversity is fundamental to the activity’s design. Activities marked significant have another primary objective but have been formulated or adjusted to contribute to biodiversity. For providers not applying the Rio marker and for TOSSD data, comparable principal-like and significant-like categories are identified using available policy markers, keywords, purpose codes, SDG tags and project descriptions. The estimate includes 100% of the value of activities where biodiversity has been tagged or identified as principal objective and 40% of the value of activities where biodiversity has been tagged or identified as a significant objective. The coefficient provides a standardised attribution assumption. For further details see the methodology description in (OECD, 2024[20]).

Commitments and disbursements

The estimates derived from CRS, which account for most of the public development finance, are based on commitments. However, some data derived from the TOSSD are based on disbursements. South-South cooperation providers also contribute non-monetised support reported to TOSSD which is not reflected in the estimates presented here.

Private finance

The global estimate of private biodiversity finance combines data on private return-seeking finance (investment) and data on private non-return-seeking finance. Private investment data derives from the report Gaining Ground: State of Private Investment in Nature report (Bennett et al., 2026[7]). Other private finance flows are identified by integrating data from the OECD CRS on philanthropic finance, OECD’s Private Philanthropy for Development database and analysis of biodiversity-focused NGO finance.

Data sources and methodology for private investment

Data in Bennett et al. (2026[7]) are based on an analysis combining three complementary data sources – survey outreach, desk research, and public data aggregation. Survey respondents represented approximately USD 207 trillion in assets under management. Seventy responses were received, of which 62 investors reported having made nature-related investments since 2020 and provided information on both financial performance and conservation outcomes. The study focuses exclusively on investments “expected to generate measurable positive impacts for biodiversity and ecosystems”, while also seeking a financial return. As a result, broader climate, ESG, or impact investments without a direct and verifiable nature objective were excluded, as were grants, philanthropy, public environmental funds, and other forms of nature finance that do not pursue financial returns. Similarly, investment vehicles that referenced nature as part of a broader mandate (e.g. climate resilience or green transition funds) were excluded where actual allocations to nature could not be confirmed. While the terms direct and indirect are not used, the scope implies both direct and indirect finance for biodiversity. The report covers and categorises data into six categories: sustainable agriculture, sustainable forestry, nature-based climate solutions, ecosystem restoration and conservation, mixed landscape portfolios, and enabling and commercialization technologies.

Survey respondents are concentrated primarily in North America and Europe, potentially resulting in the underrepresentation of nature-related investment activity in regions such as Africa and Asia due to lower participation rates and limited public disclosure. In addition, portions of the analysis relied on large language model (LLM)-supported data extraction from public sources. While extraction workflows were validated against ground-truthed samples and refined to minimise errors, some inaccuracies may remain, particularly for more nuanced variables such as expected returns, debt structures, and equity stakes. Despite these constraints, the results are considered robust because survey data were systematically triangulated with findings from desk research and public data aggregation, enabling inconsistencies to be identified and corrected. The consistency of patterns observed across all three data sources, together with alignment with broader industry reporting, provides confidence in the overall conclusions and represents a novel approach to quantifying private investment in nature.

Data sources and methodology for philanthropic funding and private income of biodiversity-focused NGOs

Philanthropic funding

Estimates of philanthropic development finance are derived from the OECD Creditor Reporting System (CRS), using the most recent data available for 2023 and 2024 and reporting a two-year average. As the CRS captures only a subset of international philanthropic flows and does not cover domestic philanthropy, these data were supplemented with information from the OECD's Private Philanthropy for Development database. This database contains nearly 120 000 philanthropic activities reported by 506 philanthropic organisations across 28 countries for the period 2020-23 and is compiled from the OECD CRS, dedicated OECD surveys and selected secondary data sources. The underlying reporting framework is aligned with OECD-DAC statistical standards and classifications, ensuring comparability with official development assistance (ODA). Further details on the dataset and methodology are available in (OECD, 2026[21]), Private Philanthropy for Development (Third Edition): Taking Stock of Philanthropy’s Contribution to Development. [oecd.org]

To avoid double counting, all cross-border philanthropic flows already reported through the CRS were removed from the OECD philanthropy database. The remaining activities – including domestic philanthropy and cross-border philanthropic flows not captured in CRS, such as some South-South philanthropy – were retained. The most recent year in the OECD philanthropy database (2023) falls within the study period and was therefore included in the analysis.

Biodiversity-related philanthropic finance was identified through a multi-stage screening process. For activities reported through the CRS, records carrying a biodiversity Rio marker were selected first; the OECD Private Philanthropy for Development database does not include this marker. In both datasets, the initial selection was supplemented using biodiversity-related DAC purpose codes, SDG 14 and SDG 15 tags, and a keyword search across all remaining activities. Project titles and descriptions were then reviewed individually, and activities were included only where biodiversity conservation, restoration, sustainable use or related ecosystem outcomes were a clear and explicit objective. Activities with only incidental, indirect or insufficiently demonstrated biodiversity relevance were excluded, including some tagged to SDG 14 or SDG 15.

The estimate includes 100% of the finance associated with activities for which biodiversity was marked or identified by the authors as a principal objective (CRS) and 40% of the finance associated with activities for which biodiversity was marked or identified by the authors as a significant objective.

Private income of biodiversity-focused NGOs

The estimate for biodiversity-focused non-governmental organisations was derived from annual financial reports for all available years during 2023-25 and focuses on private revenue rather than expenditure. Expenditure was excluded because public and philanthropic funds are often channelled through NGOs and may already be captured elsewhere in the analysis. Private revenue included donations, legacies and bequests, memberships and individual contributions, corporate contributions, commercial contracts, and own-source income such as sales of goods and services. To reduce potential double counting, the separate estimate of biodiversity-related philanthropy derived from OECD datasets was deducted from the NGO revenue total. This is a conservative adjustment: the OECD datasets cover philanthropic funding to a much wider range of recipient organisations than the eight NGOs reviewed, and the underlying recipients, years and transactions could not be matched directly. The deduction therefore likely exceeds the actual overlap and may understate the remaining private-finance estimate.

Data sources for other market indicators of biodiversity finance

Private finance mobilised by official development finance

Estimates are based on OECD data on private finance mobilised by bilateral and multilateral development finance interventions which is collated in the Creditor Reporting System (CRS). OECD mobilisation methodologies are instrument-specific and apply principles of causality and pro-rated attribution, including avoiding double counting where several official providers contribute to the same mobilisation. Biodiversity relevance is identified from the associated development-finance activities. The approach for identifying biodiversity activities is the same outlined for international public development finance above. The estimate includes the full value of activities for which biodiversity is a principal objective and 40% of the finance associated with activities where biodiversity is a significant objective. Figures are annual averages for 2023-24 commitments in constant 2024 USD.

Domestic corporate expenditure

The estimate is derived from Environmental Protection Expenditure Accounts (EPEA) for 25 EU countries and covers corporate expenditure classified under protection of biodiversity and landscapes in 2023. The value was calculated by summing compensation of employees; intermediate consumption of products other than environmental protection (EP) services; intermediate consumption of services by corporations as specialist producers; intermediate consumption of EP/resource management (RM) services; and gross fixed capital formation (GFCF) and acquisitions less disposals of non-produced non-financial assets. EPEA follows national-accounting principles and internationally standardised classifications of environmental activities, providing a relatively consistent basis for comparison across countries. The estimate is limited to expenditure captured within the EPEA accounting boundary and therefore does not represent all corporate spending or investment that may benefit biodiversity.

Bonds with biodiversity use of proceeds

Estimates are based on (UNEP, 2026[22]) analysis of BloombergNEF and Bloomberg Terminal data on green and sustainable bonds listing biodiversity among eligible uses of proceeds. Where an issuance lists several uses of proceeds, the total issuance is allocated equally across them; for example, one-tenth of a bond listing ten uses is attributed to biodiversity. This provides an indicator of potential biodiversity-related issuance rather than actual allocation The authors acknowledge that these estimates may overstate biodiversity finance, as issuance is allocated equally across listed uses of proceeds, whereas actual allocations are typically uneven and biodiversity often receives a relatively small share. One analysis of bonds issued in 2021-22 that listed biodiversity found that only 3.7% of proceeds were ultimately allocated to biodiversity projects (Bromley, 2024[23]).

Biodiversity-labelled investment funds

The estimate draws on MSCI's identification of 24 “pure-play” biodiversity-labelled funds, defined as funds that include biodiversity in both their name and investment strategy. These funds held USD 1.6 billion in assets under management as of September 2024. AUM measures the stock of assets managed by the funds and should not be treated as an annual finance flow or as finance directly reaching biodiversity projects. Investment approaches also vary, including strategies focused on reducing portfolio exposure to biodiversity risks as well as investing in companies providing solutions to biodiversity loss.

Voluntary biodiversity credits

Estimates draw on transaction data compiled by bloomlabs for the emerging voluntary biodiversity-credit market. Public information on transactions remains incomplete, particularly for privately negotiated transactions, and estimates are revised as additional data become available. The figures should therefore be interpreted as observed or tracked market activity rather than a comprehensive estimate of global biodiversity credit finance.

Water-quality trading

The estimate is drawn from Smith et al. (2025[24]), which tracks investment in nature-based solutions for water security globally. Water-quality trading enables water users to purchase credits generated through activities that reduce pollutant loads or improve water quality, often within regulatory frameworks. Transactions amounted to at least USD 21 million in 2023. Some credit-generating activities can contribute directly to biodiversity, but addressing biodiversity loss is not necessarily their primary purpose; consequently, the full value of water-quality credits is not treated as biodiversity finance.

Biodiversity-relevant voluntary carbon markets

The estimate is based on the retirement of voluntary carbon credits from project categories with a plausible basis for biodiversity relevance. Retirement volumes are drawn from the Berkeley Carbon Trading Project (Quartson et al., 2026[25]), and valued using average traded prices for the credit category in the year of retirement, from Ecosystem Marketplace. While there is usually a time lag between the trading and retirement of carbon credits and considerable price variation within credit categories, this high-level estimate based on publicly available intelligence is deemed adequate for the purposes of this report. The lower estimate includes forest and other land-use projects certified under the Climate, Community and Biodiversity Standards; the upper estimate additionally includes REDD+ projects not captured in the lower bound and relevant Gold Standard forest and land-use projects. The resulting USD 488-540 million estimate is a proxy value for biodiversity-relevant carbon market revenues rather than observed expenditure on biodiversity. Biodiversity relevance is inferred from project type and certification and does not establish the share of expenditure directed specifically to biodiversity or the biodiversity outcomes achieved. The range reflects different levels of retired credit volumes that could be considered biodiversity-relevant but does not account for potential uncertainty in the pricing of retired carbon credits and how much of the credit value reaches biodiversity-relevant activities.

Notes

← 1.  Available at Data.imf.org. Data from EU members is collected via Eurostat, for non-EU members, by the OECD, and for non-OECD members, by the IMF.

← 2.  Group 0502 includes: • the environmental rehabilitation of abandoned mining and quarrying sites and related expenditure. It excludes: • farming and gardening activities (outside the scope of CEP classification except organic farming included under CEP 0501); • the protection and rehabilitation of historic monuments or predominantly built-up landscapes; • the control of weeds for agricultural purposes; • the recreational structures and spaces such as urban parks and gardens, golf courses and other sports facilities; • management of zoos, oceanarium, aquarium, and of city greenery; • establishment and maintenance of green spaces along roads; • ‘extraction’ activities, such as ‘hunting’ or ‘fishing’; • activities related to soil cleaning (to be reported under CEP 0501).

← 3.  Since 2024, the CRS database is divided into these three pillars: ODA, OOF, and PSI. PSI captures the parts that were previously under ODA and OOF that mobilise private finance through international development finance.