Case study: carbon accounting approaches and an analysis of Meta’s 2023 data center electricity consumption and clean energy procurement

Download the case study PDF:
How carbon accounting approaches do (or don’t) reveal real-world impacts: An analysis of three methodologies to report emissions from Meta’s 2023 data center electricity consumption and clean energy procurement.

Summary

Since 2020 Meta has matched 100% of its electricity use with more than 15 gigawatts of long-term clean energy purchase commitments, making it one of the world’s largest corporate buyers of clean energy. As a result, Meta has reduced its electricity-associated emissions reported under the current industry standard, the Greenhouse Gas Protocol’s (GHGP) market-based method, to nearly zero. But how well do these standard reported methodologies capture Meta’s physical emissions in the real world?

The GHGP has played a key role in driving over 200 gigawatts of corporate clean energy purchases. But today it is undergoing a major revision — its first in over a decade. Since it was last updated, many power grid operators and third-party providers started releasing far more granular and complete emissions data than were available at the time the current system was devised.

These new data show that the carbon intensity of electricity varies substantially by time and exact location. The emissions impact of using or generating electricity depends not just on how much is consumed, but also on when and where — and what technologies (coal, natural gas, hydropower, etc.) are on the grid at that moment. These variations in emissions impact have become even more pronounced in recent years due to the widespread deployment of clean energy. In certain times and places electricity has become very clean — for example, in West Texas when the wind is blowing — while others have changed little.

If we’re serious about reducing pollution from electricity grids and power sector decarbonization, then we need to measure the emissions impact of electricity consumption and clean energy generation more accurately, enabling companies to make informed decisions about where and when clean energy investments can have the greatest impact. The GHGP revision process currently underway provides a critical opportunity to ensure this foundational global standard better reflects real-world variations in electricity’s carbon intensity across time and place.

A key element of past GHGP updates has been examining case studies. At this pivotal moment in the GHGP’s evolution, Meta engaged WattTime to analyze its 2023 data center operations and clean energy procurement using three different methodologies currently under consideration by the GHGP. The goal was to use Meta’s real-world data as a test case for the potential implications of different approaches for all companies.

The three methodologies examined in the case study were: 1) Annual Matching (current GHGP methodology), 2) Hourly Matching (24/7 CFE methodology), and 3) Carbon Matching (emissions matching methodology). This analysis strongly suggests a need for the GHGP (and other carbon accounting frameworks) to adopt more accurate carbon accounting methodologies such as Carbon Matching that more accurately reflect real-world emissions impact and empower companies to make more targeted, better informed, and higher-impact clean energy investments. Methodologies such as carbon matching are well aligned with the three main criteria of the GHGP Scope 2 revisions: scientific rigor, will drive ambition in climate action, and feasibility.

Download the case study PDF:
How carbon accounting approaches do (or don’t) reveal real-world impacts: An analysis of three methodologies to report emissions from Meta’s 2023 data center electricity consumption and clean energy procurement.