MSCI Upgraded Williams to AAA. Its Scope 1 and 2 Emissions Are Back at 2023 Levels.

MSCI upgraded Williams to AAA in March 2026, its highest ESG rating. Sustainalytics places the company in the top one per cent of the Refiners and Pipelines industry, CDP has scored it A minus for three years against an industry average of B, and it has held a place on the Dow Jones Best-in-Class North America Index for six.

 

Across the year those assessments cover, Williams emitted 16.25 million metric tons of carbon dioxide equivalent from Scope 1 and Scope 2. In 2023 the figure was also 16.25 million. In 2021 it was 14.72 million.

 

That gap runs through the company's 2025 Sustainability Report, published on 30 July 2026, whether or not the report intends it to. Williams operates more than 32,000 miles of pipeline across 24 states and handles roughly one-third of US natural gas, so small percentages meet large absolute numbers quickly. The report lands as the company positions gas as the answer to data centre demand, and as projects blocked for a decade move into construction.

 

Williams at a Glance

• More than 32,000 miles of pipeline across 24 US states
• Approximately one-third of US natural gas handled
• Approximately 60 per cent growth in energy throughput since 2018
• 28 per cent reduction in Scope 1 and 2 emissions intensity since 2018
• 123,382 metric tons of Scope 1 methane emissions in 2025

 

What intensity measures, and what it does not

 

Williams does not burn most of what it moves. Its own emissions come from compression, processing and venting, and its climate commitment is expressed as intensity: metric tons of CO2e per thousand MMBtu of throughput. Intensity improves whenever emissions grow more slowly than activity, so it can fall steeply while total emissions rise.

At Williams, both things happened: throughput grew roughly 60 per cent between 2018 and 2025 and Adjusted EBITDA grew 67 per cent, while emissions intensity fell 28 per cent to 0.81 metric tons CO2e per thousand MMBtu, close to the 30 per cent reduction targeted for 2028. The absolute totals moved the other way.

 

 

So the ratio improved by more than a quarter while the tonnage grew by a tenth. Prior-year Scope 1, carbon and methane figures were restated for acquired assets, additional emission sources and methodology updates, so the trend is measured against a moving baseline. Williams is candid about the direction of travel, stating that absolute emissions may increase with growth and that expansion lets gas displace more emissions-intensive fuels elsewhere. That holds up in US power sector data, where gas-fired generation rose over fifteen years while sector CO2 fell. It also points at emissions on somebody else's books, from a company whose own have stopped falling.

 

Scope 2 deserves separate attention, having risen 38 per cent since 2021 while methane fell. Some compressor modernisation replaces gas-fired equipment with electric motor drives, which depending on the electricity source can shift part of the emissions profile from Scope 1 to Scope 2 rather than remove it, and the report acknowledges that possibility. Nothing proves electrification caused the full increase, and more purchased power across a growing asset base is the simpler explanation. But Williams reports Scope 2 on a location-based method only, and its 14 per cent renewable electricity figure comes from EPA eGRID grid averages rather than anything the company bought.

 

Intensity shows whether Williams is operating each unit of infrastructure more efficiently. Absolute emissions show whether its reported operational footprint is shrinking.

 

Falling methane, and how it was measured

 

The methane numbers are the strongest part of the report, and unusually for this sector they hold up in absolute terms rather than only as a ratio. Scope 1 methane fell to 123,382 metric tons in 2025 from 134,451 in 2024, after peaking at 146,167 in 2023, making this a second consecutive year of decline. Methane intensity has fallen roughly 40 per cent since 2018, against a 2028 target of 0.0375 per cent, and the company beat the 5 per cent reduction built into its 2025 incentive programme, which ties part of employee pay to the result. UNEP estimates methane causes around a third of current global warming, so reductions this decade count for more.

 

The credibility of that figure rests on improvements in both measurement and operations. By the end of 2025, 143 facilities had entered its quantification, monitoring, reporting and verification programme, pairing point sensors and continuous monitoring with aerial surveys and optical gas imaging. One detail here works in Williams' favour. Inventories built on generic factors have historically underestimated real methane, so measuring properly can raise reported figures even while detection and repair improve, which is reason to treat this decline as credible.

 

Fort Lupton in Colorado shows the model working, where continuous sensors flagged abnormally high emissions after a shutdown and restart, traced to rod packing a contractor had installed incorrectly on two compressors and repaired months before the next scheduled survey would have found it. The reduction itself came from accumulated work rather than one flagship project, spanning compressor modernisation, vent gas recovery, pneumatic controller retrofits, blowdown capture, leak detection and repair, and well-site isolation. The quantified results for 2025 included:

 

• Four compressor engines replaced across two projects on Northwest Pipeline
• A 10.6 per cent fall in blowdown methane where gas was redirected or recompressed
• More than 100 metric tons of methane reduced through SlipStream vent-capture technology
• Roughly 740 of more than 1,000 identified low-return wells isolated in the Wyoming upstream assets, an estimated 80,000 metric tons of CO2e with no capital spend beyond labour

 

The expected 50 per cent methane cut at participating compressor stations by 2030 is a forecast rather than a result, as is the patent-pending Energy Exchange Compressor, which has completed shop testing but not field testing. Williams also reports achieving the OGMP 2.0 Pathway to Gold Standard for a second consecutive year, working at Level 4 and pursuing Level 5 on a third of its material assets. Pathway status means a company is on track to reach Gold Standard rather than having arrived, and UNEP's guidance for policymakers is explicit that the status is "solely based on quality of reporting", with meeting or missing a methane target not a factor. The designation assesses reporting quality and progress towards measurement-based data. It does not indicate that Williams' methane emissions are low or that its reduction target has been achieved. ERM Certification and Verification Services gave limited assurance over selected figures, narrower than an audit and covering only what was chosen for review.

 

Flaring, spills and the contractor gap

 

Several numbers move the other way, all in the performance data table and none on the highlights page:

• Gas flaring: 184,260 metric tons in 2024 to 268,920 in 2025, roughly 46 per cent higher
• Volatile organic compounds: 8,599 tons to 13,846 tons
• Hydrocarbon spills above one barrel: four to ten
• Water withdrawal: 3.66 million gallons to 434.9 million

 

Williams also missed its goal of cutting reportable spills and releases by 10 per cent, recording 161 agency-reportable incidents, though the 2024 baseline was not recalculated for assets acquired in 2025 and operations expanded into jurisdictions with lower reporting thresholds. The inclusion of acquired production assets clearly explains most of the water increase. The report does not provide an equivalent explanation for the sharp rise in VOC emissions. Two series run in Williams' favour, with nitrogen oxides down 22 per cent to 20,015 tons and the employee recordable incident rate improving to 0.59, its best in five years. Contractor figures have risen every year since 2021, from 9 recordable accidents to 28 in 2025, with the contractor rate flat at 0.61. Safety is improving for badged employees and not for the people working beside them.

 

Pipes, power and the Zamarin strategy

 

This is the first full report under Chad Zamarin, who became chief executive on 1 July 2025 after fourteen years of Alan Armstrong, having come up through corporate strategic development with a remit covering New Energy Ventures, upstream and acquisitions. The growth engine is now electricity. Williams has announced roughly 2.6 GW of onsite gas generation for large-load customers in grid-constrained regions, arguing these projects deliver power faster than grid interconnection and can incorporate battery storage, renewables or carbon capture. Zamarin writes that Williams is "delivering clean and reliable power where it's needed without burdening local grids".

 

Independent data confirms the demand and complicates the emissions picture. IEA analysis finds gas is already the largest single source of electricity for US data centres at more than 40 per cent, and expects gas and coal together to meet over 40 per cent of additional demand from those facilities through 2030. Generation added to serve that demand still produces combustion emissions at the point of use.

 

The rest of the expansion runs the same direction, adding 1.8 Bcf/d of gathering capacity, 1.1 Bcf/d of transmission with 4.4 Bcf/d announced, and a minority interest in Louisiana LNG. Each move enlarges the denominator in the intensity ratio, which is how the target is built. An intensity goal can be met by cutting the numerator, growing the denominator, or both at once.

 

Permits, Scope 3 and what the ratings measure

 

NESE secured its Clean Water Act permits from New York and New Jersey in November 2025 after years of state-level refusal, and broke ground in Brooklyn on 14 April 2026 with the EPA administrator, the interior secretary and the energy secretary present. Constitution Pipeline, cancelled in 2020, returned to FERC through a petition seeking reissuance of its previous certificate authority. The report contains sections on stakeholder relations, environmental justice and overburdened communities. Set those beside a permitting outcome resolved at federal executive level over sustained local objection, and the distance between the two accounts of the same process becomes hard to ignore.

 

Three disclosure choices are underneath this, each reasonable alone and less so together. Scope 3 emissions are not quantified in the report, with the GRI index directing readers to Williams' CDP questionnaire. Net zero by 2050 is described as an ambition rather than a target. And the intensity commitment is not SBTi-validated, the report saying only that it was informed by the SBTi's Target Setting Manual, a carefully constructed sentence.

 

The ratings do not all measure the same thing. MSCI's AAA is industry-relative, assessing how Williams manages financially relevant ESG risks and opportunities against its peers. Sustainalytics measures unmanaged ESG risk to enterprise value, while CDP scores reported environmental disclosure, management and action. S&P Global's assessment examines industry-specific sustainability practices and performance. None is designed as a direct test of whether Williams' absolute Scope 1 and 2 emissions are falling.

 

ESG ratings can reward strong governance, disclosure and methane control without answering whether a company's growth strategy is compatible with falling economy-wide emissions.

 

What Williams has proved, and what it has not

 

Williams has shown that a large gas infrastructure business can cut methane in absolute terms while growing, through two consecutive years of decline, measurement that catches faults between surveys, and a compensation link that puts the target in front of employees. The other half is unproven. Scope 1 and 2 emissions in 2025 were identical to 2023 and higher than 2021, Scope 2 has risen every year since 2021, and flaring rose sharply. Operational improvement has not yet outrun the emissions consequences of an expansion that is accelerating.

 

Six things would change that, and most should show up within two reporting cycles:

• A sustained fall in absolute Scope 1 rather than a plateau
• Scope 2 falling, ideally with market-based reporting alongside the location-based figure
• Continued absolute methane reductions rather than intensity alone
• Delivery on the 0.0375 per cent methane intensity target for 2028, and OGMP Level 5 across more material assets
• Separate emissions reporting for the onsite power projects, and carbon capture operating rather than under evaluation
• Better spill and flaring performance on a like-for-like boundary

 

Future reports will provide a harder test as Rimrock and Saber contribute a full reporting year and Williams begins commissioning its onsite power projects. A sustained decline in absolute emissions across that larger operating base would strengthen Williams' position. Continued flat or rising totals would show how much of the reported progress still depends on intensity.

 

Sources

Williams 2025 Sustainability Report, 30 July 2026 (performance data table pp. 95-97; emissions pp. 35-42; CEO letter pp. 5-6; assurance statement pp. 108-122). Ratings as reported by Williams: MSCI, March 2026; Sustainalytics, May 2026; CDP 2025 Corporate Questionnaire.

UNEP International Methane Emissions Observatory, An Eye on Methane and A Blueprint for MRV, 2025. IEA, Energy and AI (2025) and Key Questions on Energy and AI (2026). US EPA release on the NESE construction launch, 14 April 2026.