Breakthrough in Belém: towards implementing the Paris Agreement

I. Recent evolution of global emissions

The lesson from more than three decades of multilateral efforts to curb global warming — and a decade of targets under the Paris Agreement — is unambiguous. The global economy has been quietly becoming less carbon-intensive, yet greenhouse gas emissions continue to rise.

There is no contradiction here. This is the central challenge of the climate transition in a world of rising living standards: decarbonisation is under way, but economic growth is still outpacing it. Climate outcomes depend not on whether decarbonisation is happening, but on whether it proceeds faster than the combined growth of population and income per capita.

The arithmetic is simple and relentless. The growth rate of emissions equals the sum of GDP growth and the change in carbon intensity, defined as emissions per unit of output. If GDP grows faster than carbon intensity falls, emissions rise.

That is precisely what has happened since 1990, as shown in the table below. Global emissions have grown more slowly than output, largely thanks to improvements in energy efficiency — but nowhere near fast enough to stabilise temperatures at the levels envisaged in the Paris Agreement.

World: average annual geometric growth rates, 1990–2023 (% per year)

Population	GDP	GDP per capita	Emissions (CO2e)
1.26	3.00	1.74	2.1

The geographical pattern of carbon emissions

Behind the global average lie three sharply different economic and emissions regimes: China; advanced economies; and emerging and developing market economies excluding China (EMDEs). These are illustrated in the table below. [Note: figures are approximate estimates based on aggregations from multiple data sources.]   

Average annual geometric growth rates, 1990–2023 (% per year)

	Population	GDP	Emissions (CO2e)
Advanced economies	0.35	1.7	0.25
China	0.50	8.0	5.8
EMDEs	1.20	3.8	2.0

No country has shaped the recent trajectory of global carbon emissions more than China. Since 1990, its economy has expanded at close to double-digit annual rates, while emissions have grown at roughly 6 per cent a year. The gap between these figures implies an extraordinary annual decline of more than 2 per cent in carbon intensity.

China has therefore been decarbonising rapidly. Energy efficiency has improved markedly, and the power mix has diversified at speed. Energy use per unit of output has fallen sharply. Yet scale overwhelms efficiency: the outcome is cleaner growth but still rising emissions.

Advanced economies tell a different story. Since 1990, their combined GDP has grown by around 1.75 per cent a year, while territorial emissions have edged down by roughly 0.25 per cent annually. This reflects familiar forces: slow population growth, deindustrialisation, gains in energy efficiency and the gradual substitution away from coal. But there is an important caveat. Part of the decline reflects the relocation of carbon-intensive industries to poorer countries — notably China — rather than their elimination. Carbon has been displaced geographically as much as it has been destroyed technologically.

The real challenge lies in the EMDEs. Taken together, these economies have grown rapidly – between 3 and 4 per cent a year since 1990. Their emissions have increased more slowly, implying the fastest proportional decline in carbon intensity of any bloc, averaging above 2 per cent per year.

This is hardly surprising. Historically, latecomer economies tend to adopt more efficient technologies as they invest in new capital. Their power systems incorporated a higher share of natural gas and renewables earlier, and a younger capital stock embeds higher energy efficiency.

Yet absolute emissions continue to accelerate. Populations grow faster than in the rich world. Urbanisation intensifies. Industrial demand rises. Like China, these countries are decarbonising rapidly in relative terms, but in absolute terms they are still pushing global emissions higher.

Forests and global carbon emissions

Another dimension of emissions accounting is the net absorption of CO₂ by forests. The figures above cover anthropogenic emissions — mainly from energy, industry and transport — but exclude those from land use, land-use change and forestry (LULUCF), driven by deforestation and partially offset by forest regeneration.

These historical series are imperfect, but their order of magnitude is widely accepted: around 10 per cent of global emissions, and on a declining trend. The latter reflects long-term forest regrowth in advanced economies and the declining role of wood as a primary energy source in the developing world.

As with fossil-fuel emissions, LULUCF tells three distinct stories. In advanced economies it is a net sink; in China it is close to balance or slightly negative; in EMDEs it is unequivocally a net source, dominated by deforestation in tropical forests.

II. A climate strategy for the next decade

Three core policy implications emerge from this diagnosis of the sources and composition of emissions growth.

First, energy efficiency alone will no longer sustain the transition. Marginal gains will be smaller and more costly. The substitution of fossil fuels with clean electricity must now drive the global decline in carbon intensity.

Second, net deforestation — though lower than in the past — remains a significant source of emissions and a vital threat to planetary life. Protecting forests, especially tropical ones, must be a central pillar of climate strategy.

Third — and most importantly — the geographical centre of future emissions has shifted. Climate outcomes will be determined far less in advanced economies and far more in the developing world.

China, however, has crossed a structural threshold. With high domestic savings, industrial scale and technological capability, it can continue to decarbonise largely under its own momentum.

The EMDE bloc is different. It will account for most of global population growth and a substantial share of new, climate-relevant infrastructure investment. This will require financing capable of sharply reducing carbon intensity while simultaneously addressing rising adaptation needs.

Despite recent innovations in climate finance, domestic savings in these countries fall far short of what is required. By 2035, annual external financing needs for clean energy, transition and adaptation are expected to reach $1.3tn, according to a report by the Independent High-Level Expert Group (IHLEG), led by Nicholas Stern and endorsed by the UN in Baku.

The same IHLEG figures make clear that foreign public resources are wholly insufficient. Multilateral development banks face capital constraints. Official aid from rich countries has stagnated. And amid the seismic geopolitical shifts of the century, climate has slipped down the political agenda in advanced economies. The inescapable conclusion is that private foreign capital will have to play a far larger role.

This defines the third priority of climate policy: increasing international private flows to EMDEs by no less than sixteen-fold over the next decade, covering at least half of the $1.3tn requirement.

III. International capital markets and the limits of private finance

Reality is less encouraging. Recent research shows that developing countries’ access to large-scale capital markets is limited and highly uneven. Only a narrow group of EMDEs with credit ratings close to investment grade can raise significant volumes of finance.

These countries can attract foreign capital through well-designed national programmes and risk-mitigation schemes. Poorer countries – typically far below investment grade or not even rated – face the opposite reality. The same is true for most adaptation projects, which generate little or no private revenue. In such cases, concessional public or multilateral finance does not complement the market; it is a precondition for finance to exist at all.

The solution proposed by experts is targeted risk-sharing. Donor governments and multilateral banks must deploy their scarce capital to reduce private risk, using loss-absorption layers calibrated to country- and project-specific risks.

In the larger emerging economies, blended-finance schemes already exist and could be scaled up – such as Brazil’s EcoInvest or the IDB’s Reinvest+. These countries, generally larger, account for no less than two-thirds of emissions in the developing world and are therefore pivotal.

For most low-income countries, however, the required instruments are more complex: multilateral financial-intermediation funds dependent on sovereign contributions from rich countries. However, given current fiscal conditions in many of those economies, politically feasible increases on the scale required to mobilise sufficient private capital appear unlikely.

The conclusion of the study is blunt. Even the most sophisticated de-risking mechanisms will not suffice. Financing the transition in roughly one hundred countries without access to deep international capital markets will require substantial new sources of public external finance — special levies, new IMF Special Drawing Rights allocations, the redirection of fossil-fuel subsidies — and, above all, an integrated global carbon market.

The importance of reviving carbon markets

International carbon credits were central to North–South climate finance under the Kyoto Protocol’s Clean Development Mechanism. Its successor — Article 6 of the Paris Agreement — took years to become operational, but the impasse has finally begun to ease.

Decisions taken in Baku to accelerate the implementation of Article 6, combined with the rapid spread of national emissions-trading systems, point to a potential revival of global carbon markets.

This momentum has been reinforced by Brazil’s proposal to create an Open Global Coalition to harmonise standards across national markets. The initiative has gained support from key jurisdictions, including the EU and China. If integrated into Article 6, it offers a plausible path towards an international carbon market capable of transforming climate finance.

This matters because carbon pricing directly raises returns on transition projects in EMDEs. By improving profitability, it magnifies the impact of blended-finance instruments, lowers the cost of capital and unlocks far larger investment flows.

IV. The legacy of COP30: the three roads of Belém

Debates took place in Belém under the Brazilian presidency explicit guidance to make COP30 the “COP of implementation”. The premise was that the transition is urgent and that the obstacles no longer lie in technology or multilateral institutional architecture, but in the political will of the UNFCCC’s major coalitions and in the inherent difficulties of consensus-based decision-making.

The official closing communiqué announced a “political breakthrough”, launching three parallel processes to be steered by the Brazilian presidency through 2026, aligned with the core themes outlined above, namely:

1. A debate on transitioning to a fossil-fuel-free economy in a just, orderly and equitable manner.
2. A Forests and Climate Roadmap to halt and reverse deforestation.
3. An integrated package to accelerate national mitigation and adaptation plans in developing economies, underpinning climate-finance demands from EMDEs — already a central item on the Belém agenda — towards the $1.3tn goal.

This third and crucial mandate was formalised in the so-called “Mutirão Decision”, a public document consolidating a set of decisions under what is termed the Global Implementation Accelerator.

The process will establish a two-year work programme on climate finance, including Article 9.1 of the Paris Agreement, which requires developed countries to provide financial resources to support developing countries in mitigation and adaptation, in continuity with commitments under the Convention.

Once completed ahead of COP31 in Antalya, Turkey, this work is intended — according to the Decision — to serve as a roadmap to achieve the US$1.3 tn annual target, concluding the unfinished business of the Baku-to-Belém Roadmap. The second act of Brazil’s presidency now begins.