The world's largest pension funds: what they own, ESG and fossil fuels
The world's biggest pension funds do not all treat ESG the same way. Compare GPFG, GPIF, NPS, CPPIB, ABP, CalPERS and PFZW.
Financial information only
This article is for informational and educational purposes only. It is not financial advice, investment advice, pension advice, tax advice, a recommendation, or a personal financial promotion. Pension and investment values can rise and fall, and you may get back less than you invest. Speak to a financial adviser authorised by the FCA (Financial Conduct Authority) before making pension or investment decisions.
The world's biggest pension and retirement-linked funds do not all treat ESG (environmental, social and governance) in the same way. Some exclude companies. Some keep holdings and engage. Some use ESG indices. Some allocate capital directly to climate solutions. The differences matter because these institutions shape markets, company behaviour and the evidence ordinary pension savers should ask from their own providers.
For practical pension checks, start with our green pensions guide and green pension funds UK guide. For related institutional context, compare this article with our explainers on Norway's sovereign wealth fund, BlackRock, climate risk in investment portfolios and sustainable funds.
The short answer
There is no single pension fund ESG model. Large pension and retirement-linked asset owners have different mandates, legal duties, political constraints, member bases and investment structures. That is why their climate and fossil-fuel approaches can look very different even when they all describe themselves as long-term responsible investors.
Norway's Government Pension Fund Global (GPFG) is known for public exclusions and detailed reporting. Japan's Government Pension Investment Fund (GPIF) uses ESG indices and stewardship across a huge passive portfolio. ABP in the Netherlands chose fossil-fuel producer divestment, then had to manage the implementation details. CalPERS in California focuses on climate-solution allocation and engagement rather than a broad fossil-fuel exclusion. CPP Investments in Canada frames climate as part of long-term value creation and the whole-economy transition. PFZW in the Netherlands is one of the clearest examples of a large pension fund with a sustainability-first public posture.
The lesson for ordinary savers is not that one model is automatically best. The useful question is narrower: what does your pension own, what claim is being made, what evidence supports it, and what happens when a company does not change?
Data note
Asset figures in this article are shown in each fund's reporting currency where possible. Approximate US-dollar comparisons can move materially with exchange rates, so the date and source matter more than a rounded dollar equivalent.
How big is pension capital?
Pension capital is one of the largest pools of long-term investment in the world.
The Thinking Ahead Institute's Global Pension Assets Study 2026 estimated pension assets across 22 major pension markets at about US$68.3 trillion at the end of 2025. Its separate ranking of the world's largest pension funds put the top 300 funds at about US$24.4 trillion.
Those are not the same number. The US$68.3 trillion figure describes major pension markets as a whole. The top 300 figure describes the largest individual funds. The funds profiled below sit near the top of the global retirement-linked asset-owner landscape, but they are still only part of the wider pension system.
The reason the scale matters is that pension capital is patient. A pension fund does not invest for next quarter. It invests across decades. That makes it structurally important for financing long-term transitions, including energy infrastructure, low-carbon industry, grid upgrades, resilient property and public-market stewardship. It also means climate and ESG decisions made today can shape portfolios for a generation.
What counts as a pension fund?
Not every institution in this article is a pension fund in the ordinary workplace sense. That distinction matters.
Occupational pension funds collect contributions from employers and employees, invest on behalf of members and pay retirement income. ABP in the Netherlands and CalPERS in the United States are examples, although their structures and legal environments differ.
Public pension reserve funds hold assets that support national pension systems. Japan's GPIF manages reserves for Japan's national pension system, but it is not a personal account provider in the way a workplace pension scheme is.
Sovereign funds with retirement-linked mandates are built from state revenues rather than member contributions. Norway's GPFG is the most important example in this article. It carries the word "pension" in its name and appears in pension-fund rankings, but it is really a sovereign savings vehicle built from petroleum revenues.
What counts as a pension fund?
Norway's GPFG is included because it is one of the most important publicly managed long-term capital pools in the world and because its responsible-investment model is widely studied. It should not be confused with an ordinary workplace pension scheme. For a full explanation, read Norway's sovereign wealth fund explained.
Seven major funds at a glance
The table below compares seven of the largest and most influential retirement-linked asset owners. It is not a strict global top-seven ranking. PFZW is included because of its importance to sustainable pension practice, not because it is necessarily seventh by assets.
| Fund | Latest checked assets | Date and source basis | Institution type | ESG or climate model | Fossil-fuel stance |
|---|---|---|---|---|---|
| GPFG, Norway | NOK 21,268bn | 31 Dec 2025, NBIM annual report | Sovereign fund | Exclusions, stewardship and public accountability | Public exclusion list; coal and conduct criteria; not a fossil-free fund |
| GPIF, Japan | JPY 257.4176tn | 31 Mar 2025, GPIF FY2024 annual report | Public pension reserve | ESG index integration and stewardship | No Norway-style public exclusion list; universal-owner approach |
| NPS, South Korea | KRW 1,610.4tn | End-Feb 2026, NPS official overview | National pension fund | Responsible investment programme and portfolio diversification | Growing ESG integration; fossil-fuel policy should be assessed against current fund documents |
| CPP Investments, Canada | C$793.3bn | 31 Mar 2026, fiscal 2026 result announcement | Public pension investment manager | Sustainability integration and whole-economy transition framing | Engagement over blanket divestment; invests across transition and conventional sectors |
| ABP, Netherlands | EUR 531bn | 31 Dec 2025, ABP investments page | Occupational pension fund | Divestment plus engagement | Listed fossil-fuel producer holdings sold; less liquid exposure has implementation complexity |
| CalPERS, United States | US$556.2bn | 30 Jun 2025, preliminary FY2024-25 results | Public employees pension fund | Climate-solution allocation and stewardship | No broad fossil-fuel exclusion; at least US$100bn climate-solutions target by 2030 |
| PFZW, Netherlands | EUR 252.0bn | 31 Dec 2025, PGGM/PFZW reporting | Occupational pension fund | Sustainability-focused investment policy | Exclusions, sustainability allocation and carbon-footprint reporting; use latest annual report for exact thresholds |
Four ESG models and what they mean
Before comparing the funds, it helps to separate the models. Most large funds combine more than one.
| Model | What it means | Typical strength | Typical limitation |
|---|---|---|---|
| Exclusions and divestment | Selling or avoiding holdings in specific companies or sectors. | Creates a clear public line and reduces exposure to defined activities. | Implementation can be difficult, especially for indirect or illiquid holdings; critics argue it transfers ownership rather than changing behaviour. |
| Engagement and stewardship | Keeping holdings and using ownership rights to push companies to change. | Uses voting power, dialogue and escalation while retaining influence. | Outcomes are harder to prove, can be slow, and depend on credible escalation if companies ignore investors. |
| ESG index integration | Tilting or screening portfolios through ESG-rated indices. | Works at very large scale and can be implemented through passive portfolios. | Best-in-class approaches can still hold oil, gas, mining or high-emitting companies if they score better than peers. |
| Climate-solution allocation | Committing capital to clean energy, green infrastructure, sustainable real estate or transition assets. | Creates direct exposure to the climate transition and signals allocation direction. | The climate-solutions bucket can sit beside conventional holdings elsewhere in the portfolio. |
The disagreement between these models is not just philosophical. It reflects genuine uncertainty about what works, different legal duties, different member expectations and different levels of tolerance for tracking-error, political risk and implementation complexity.
GPFG: Norway and the exclusion framework
Norway's Government Pension Fund Global had a value of NOK 21,268bn at the end of 2025, according to Norges Bank Investment Management (NBIM). Depending on exchange rates, that places it around the US$2tn mark.
The fund was built from petroleum revenues and is managed by NBIM under a mandate set by the Norwegian Ministry of Finance. Its objective is to generate long-term returns for current and future Norwegian generations.
The fund is often discussed in sustainable investment because it has a formal ethical framework. A Council on Ethics recommends exclusions or observation for companies based on defined product and conduct criteria. Product criteria include tobacco, certain weapons and coal-related activities. Conduct criteria include serious human rights violations, severe environmental damage, gross corruption and other serious ethical breaches.
This is not the same as being a clean-energy fund. GPFG remains a broad global investor. It holds stakes in thousands of listed companies, plus bonds, real estate and a small allocation to unlisted renewable energy infrastructure. Its ESG significance lies in the transparency of its rules, exclusions, voting and public reporting.
The fund's performance is also frequently misused in sustainability debates. Norway's fund has delivered strong long-term returns while using ethical exclusions. That shows exclusions do not automatically prevent a large diversified fund from participating in broad market returns. It does not prove that ESG always outperforms, or that exclusions caused the return record.
The political context matters too. In 2025, the fund's ethics and divestment processes attracted political pressure around specific exclusion decisions. That does not erase the framework, but it shows that even a highly transparent sovereign fund operates inside real-world politics.
GPIF: Japan and the universal-owner model
Japan's Government Pension Investment Fund reported JPY 257.4176tn in pension reserve fund assets under management at 31 March 2025. It is one of the largest pools of retirement-linked capital in the world.
GPIF's mandate is to contribute to the stability of Japan's public pension system. That mandate shapes the whole investment approach. It invests across domestic bonds, foreign bonds, domestic equities and foreign equities, largely through passive and externally managed portfolios.
At that scale, GPIF cannot behave like a small ethical fund that simply avoids broad parts of the market. Its ESG approach is closer to a universal-owner model: broad market exposure, ESG index selection, stewardship expectations and attempts to improve the quality of market-wide sustainability information.
GPIF has allocated significant equity exposure to ESG indices, including Japanese and global ESG index strategies. This does not make the portfolio fossil-free. It means parts of the portfolio are tilted or selected according to ESG index rules while the fund remains a market-wide investor.
The contrast with Norway is useful. Norway publishes a list of companies excluded under ethical guidelines. GPIF tries to use size, index design, manager expectations and stewardship to improve standards across the market. Neither approach is automatically superior. They reflect different mandates and different theories of change.
NPS: South Korea and a large reserve system
South Korea's National Pension Service is another giant public retirement asset owner. Its 2024 National Pension Fund Report showed KRW 1,212.9tn at 31 December 2024, while the official NPS overview page showed KRW 1,610.4tn at end-February 2026. The difference underlines why dated local-currency figures are essential for funds of this size.
NPS is important because it shows a different institutional context again. It is a national pension fund with a large and diversifying portfolio, and it has been developing responsible investment policies over time. But it is not as publicly associated with a single ESG model as Norway's exclusion framework or Japan's universal-owner index strategy.
For readers, the useful lesson is that large pension systems can be institutionally important even when their sustainability story is less simple. Some funds publish clear exclusion lists. Some publish climate targets. Some publish stewardship expectations. Others move more gradually through responsible investment guidelines, manager selection and asset-allocation changes.
That is why headline scale should never be treated as a sustainability claim. A trillion-scale pension reserve can be financially important without being easy to classify as green, fossil-free or climate-aligned.
CPP Investments: Canada and the engagement model
CPP Investments reported C$793.3bn in net assets at 31 March 2026. It manages assets for the Canada Pension Plan and is widely regarded as one of the most sophisticated public pension investment managers in the world.
CPP Investments previously announced a net-zero 2050 commitment, but its current public sustainability language is more cautious. It now places more emphasis on sustainability integration, long-term value creation and financing the real economy's transition than on blanket divestment.
That makes CPP Investments a useful case study in how institutional climate language can evolve. Rather than presenting climate as a simple portfolio-purity test, CPPIB frames sustainability as part of long-term investment management across the whole economy.
The engagement logic is familiar: selling a stake in a high-emitting company may transfer ownership to another investor, while engagement can use ownership rights to push for better disclosure, transition planning and capital discipline. The weakness is equally familiar: engagement needs evidence. If a company does not change, the fund needs a credible escalation policy or the engagement claim becomes hard to assess.
CPPIB should therefore be described as an engagement and whole-economy-transition case, not as a simple "net zero fund" case.
ABP: the Netherlands and the limits of divestment implementation
ABP is the pension fund for Dutch public sector workers, including civil servants, teachers and police officers. Its English investments page reported EUR 531bn of invested assets at 31 December 2025.
ABP is one of the most important pension-fund examples in climate investing because it announced a major fossil-fuel producer exit in 2021. The announcement covered fossil fuel producers and reflected ABP's view that continuing to invest in those companies was no longer consistent with its sustainable-economy ambitions.
The important point is implementation. Selling listed equity holdings in large fossil fuel producers is one thing. Working through private markets, indirect exposure, legacy assets and less liquid holdings is another. ABP's own public updates describe a process of selling major listed fossil-fuel producer investments while continuing to manage the wind-down of other exposures.
The better lesson is that divestment at scale has operational edges. Public announcements are easy to understand. The asset-level mechanics are more complicated.
ABP also shows the influence of member pressure. Dutch pension funds operate in a public environment where climate, fossil fuels, biodiversity and responsible investment are not abstract branding issues. Members, civil society and policymakers pay attention.
CalPERS: the United States and the political context
The California Public Employees' Retirement System reported US$556.2bn in assets at 30 June 2025 in its preliminary financial-year result announcement. It is the largest public pension fund in the United States.
CalPERS is useful because it operates in one of the most politically contested ESG environments in the world. In the United States, some states and political actors have challenged ESG investing on the grounds that it may conflict with fiduciary duty or impose values-based criteria on public assets. California has taken a different route.
CalPERS frames climate change as a financial risk and an investment opportunity. It has announced a plan to increase climate-solutions investments to at least US$100bn by 2030. It has also reported climate-solution investments near US$60bn as of 30 June 2025.
That does not mean CalPERS is fossil-free. It has not applied a broad oil and gas exclusion. Its public approach combines stewardship, risk management, climate-solution allocation and selective exclusions such as thermal coal divestment.
The CalPERS example matters because it shows one of the central legal arguments in climate investing: if climate change affects asset values, insurance costs, infrastructure risk, energy markets and corporate profitability, then considering it can be part of fiduciary risk management rather than a separate values screen.
PFZW: a large Dutch fund with a sustainability-first posture
PFZW, the Dutch pension fund for healthcare and welfare workers, is smaller than GPFG, GPIF, NPS and CalPERS, but it is influential in sustainable investment debates. PGGM's 2025 reporting showed EUR 252.0bn of PFZW assets managed at 31 December 2025.
PFZW is included because it gives the article a second Dutch comparison alongside ABP. It is not presented as the seventh-largest fund globally. It is presented as a major occupational pension fund with a prominent sustainability framework.
PFZW's public investment approach includes exclusions, sustainability allocations and carbon-footprint reporting. Its annual-report summary also discusses the share of assets linked to Sustainable Development Goals and carbon-footprint reduction metrics across parts of the portfolio.
The exact fossil-fuel thresholds, carbon-footprint baseline and asset-class coverage can change between reporting periods. That is why readers should use PFZW's latest annual report rather than relying on older summaries or a single headline carbon-reduction figure.
What this means for UK pension savers
The funds above operate at a scale ordinary savers will never copy. But they still make the pension question easier to understand.
Most UK workers are enrolled in a workplace pension. Many are invested in a default fund chosen by the scheme or provider. The sustainability of that pension does not depend on the provider's homepage language. It depends on the fund's holdings, benchmarks, exclusions, stewardship, climate reporting, risk controls and costs.
Start with the default fund. Find the fund name, factsheet, benchmark, top holdings and charges. Then ask whether the fund uses exclusions, ESG integration, climate tilts, stewardship, low-carbon benchmarks or a sustainability objective. Those are different mechanisms.
Then look for evidence. Large UK pension schemes are required to publish climate-related reporting under rules influenced by the Task Force on Climate-related Financial Disclosures (TCFD). Some schemes also publish implementation statements showing how they have acted on stewardship policies, including voting and engagement.
If the provider offers lower-carbon, ethical or sustainable self-select funds, compare them carefully. A greener-sounding option can still be more expensive, more concentrated, more volatile or less suitable for a retirement plan. The issue is not just whether a fund is greener. It is whether the fund is understandable, diversified, fairly priced and aligned with the saver needs.
| Question | What to ask for | Why it matters |
|---|---|---|
| What does my pension own? | Fund factsheet, top holdings and full holdings where available. | A provider-level climate claim may not describe the fund you actually hold. |
| Is it fossil-free? | Exclusion policy, fossil-fuel revenue thresholds and sector exposure. | ESG integration and fossil-free investing are not the same thing. |
| Does engagement have teeth? | Voting record, engagement examples and escalation policy. | Stewardship claims need evidence, not just intent. |
| Is climate risk measured? | Financed emissions, carbon intensity, scenario analysis and TCFD-aligned reporting. | Climate risk can affect equities, bonds, property, infrastructure and insurance exposure. |
| Are there alternatives? | Self-select funds, ethical options, lower-carbon funds and fee comparisons. | A better-looking option may change risk, cost or retirement-pathway assumptions. |
For a practical step-by-step version of those checks, use our green pension funds UK guide and sustainable pension transfer checklist. For fund-level evidence, use the sustainable fund factsheet checklist and FCA SDR labels explainer.
What not to conclude
The data in this article can be misused if the conclusions are too neat.
Do not conclude that large pension funds are automatically sustainable. A fund can be huge, professionally run and long-term without being low carbon or fossil-free.
Do not conclude that divestment is always better than engagement. Selling fossil-fuel holdings can create a clear public line, but the climate outcome depends on what happens next. Ownership may simply pass to someone else.
Do not conclude that engagement is always better than divestment. Engagement needs evidence and escalation. If a company ignores investors for years, a stewardship claim can become weak.
Do not conclude that ESG index integration means fossil-free. Best-in-class ESG indices can keep oil and gas companies if those companies score better than sector peers.
Do not conclude that a climate-solutions allocation describes the whole fund. CalPERS can have a large climate-solutions target while still holding broad-market exposure elsewhere.
Do not conclude that strong returns prove ESG outperformance. Performance attribution is complex. Asset allocation, market cycles, currency and sector exposure all matter.
Frequently asked questions
What is the largest pension fund in the world?
Japan's GPIF and Norway's GPFG are usually at or near the top of global pension and retirement-linked fund rankings, depending on currency moves and reporting dates. GPIF reported JPY 257.4176tn at 31 March 2025. Norway's GPFG reported NOK 21,268bn at 31 December 2025.
Is Norway's oil fund really a pension fund?
It carries the name Government Pension Fund Global and appears in pension-fund rankings, but it is a sovereign investment fund built from petroleum revenues. It does not work like a workplace pension scheme funded by employer and employee contributions.
Do pension funds invest in fossil fuels?
Many large pension funds have some fossil-fuel exposure, either directly through company holdings or indirectly through broad market indices, private assets or infrastructure. The level of exposure depends on exclusions, benchmarks, mandates and asset classes.
Can a pension fund be ESG and still own oil and gas companies?
Yes. ESG integration does not automatically mean fossil-fuel exclusion. A fund may hold oil and gas companies because it uses best-in-class ESG scoring, because it follows a broad index, or because it believes engagement is more effective than divestment.
What is the difference between divestment and engagement?
Divestment means selling or avoiding holdings in certain companies or sectors. Engagement means keeping holdings and using voting, dialogue and escalation to push companies to change. Both models are used by major pension funds. Both have strengths and weaknesses.
How can I check what my pension owns?
Look for the fund factsheet, default fund name, top holdings, full holdings if published, benchmark, charges, climate report and stewardship report. If the provider does not publish enough information, ask for it before relying on any sustainability claim.
Are green pension funds safer than ordinary pension funds?
No. Sustainable or ESG-focused pension investments still carry market risk, concentration risk, currency risk, manager risk and retirement-planning risk. Sustainability characteristics do not guarantee better returns or lower losses.
Useful source links
- Thinking Ahead Institute: Global Pension Assets Study 2026
- Thinking Ahead Institute: The world's largest pension funds 2025
- NBIM: the fund's market value
- NBIM: Annual Report 2025
- GPIF: FY2024 annual report
- GPIF: ESG investment
- National Pension Service: fund overview
- National Pension Service: 2024 National Pension Fund Report
- CPP Investments: fiscal 2026 net assets announcement
- CPP Investments: approach to sustainability
- ABP: investments
- ABP: fossil-fuel producer exit announcement
- PFZW: annual reports
- PFZW: Annual Report Summary 2025
- CalPERS: preliminary 2024-25 return and assets
- CalPERS: sustainable investments programme