英文【OECD】气候俱乐部成员钢铁和水泥行业的碳强度指标：2023-24年工作计划支柱1，模块1“克服排放强度指标的挑战”

Summary of Carbon Intensity Metrics in the Steel and Cement Sectors of Climate Club Members

Core Content

This report, prepared by the OECD for the Climate Club, explores the challenges and opportunities in computing product-level carbon intensity metrics for the steel and cement sectors. It aims to support the development of transparent, comparable, and actionable carbon intensity data to inform policy design and implementation in these industries, which are key contributors to global greenhouse gas emissions.

Main Approaches and Challenges

Scope and Boundaries

• Product-level carbon intensity metrics are often calculated using life cycle assessment (LCA) methodologies.
  • Attributional LCA (ALCA): Assigns observed emissions to a specific product and is more practical for a wide range of products due to its straightforward framework.
  • Consequential LCA (CLCA): Incorporates economic models to estimate emissions and account for feedback effects but is more complex and uncertain.
• Scope 1 and 2 emissions are typically included in LCA, while downstream Scope 3 emissions are often excluded due to data limitations and complexity.
• The temporal scope of assessments can influence carbon intensity metrics, as emission intensities may vary over time. Annual data is commonly used to smooth out variations, though hourly data is becoming more relevant in policy design.

Trade-offs Between Calculation Methods

• There are three main methods for computing carbon intensity metrics:
  • Spend-based methods: Multiply company expenditures by emission-intensity factors derived from environmentally extended input-output models.
  • Average data methods: Use industry averages for emission factors, multiplying them with activity data (e.g., material weight, fuel consumption).
  • Primary data methods: Utilize directly measured or estimated emissions at the facility level, which may be shared across supply chains.
• Accuracy vs. resource requirements is a key trade-off. More accurate methods are resource-intensive and may not be feasible for all firms, especially SMEs or those in developing countries.
• Allocation rules are essential for primary data methods, as facilities often produce multiple products. These rules help distribute emissions across different products.

Verification and Data Sharing

• Verification of emissions data remains a challenge, especially at the installation level. It is not yet a standard practice and is often optional.
• Data sharing along supply chains is complicated by fragmented reporting standards and the need for consistent verification processes.
• Initiatives are emerging to address these issues, including private and public efforts to standardize data collection and verification at the firm level.

Key Initiatives and Data Sources

Industry Initiatives

• Industry associations like the Global Cement and Concrete Association and the World Steel Association are developing guidelines and standards for carbon intensity metrics.
• Companies and research institutes are also contributing to data collection and emission accounting frameworks.

Data Sources

• The report draws on Climate TRACE and CRU data to analyze carbon intensity trends.
• Key findings:
  • The global average carbon intensity of steel has declined by 10% since 2006, but absolute emissions have increased due to higher production.
  • The carbon intensity of cement has remained stable over the past two decades.
  • Climate Club members have lower and faster declining carbon intensities in steel compared to non-members, while cement intensities are comparable.
  • Installation-level variations are significant, with power inputs being a critical factor in both sectors.
  • Production technology (e.g., EAF vs. BOF in steel) and the clinker-to-cement ratio influence carbon intensity in cement.

Lessons and Recommendations

Comparability of Data

• Country-level trends are generally consistent across data sources, but installation-level discrepancies are more pronounced.
• Methodological differences, such as emission scopes and reliance on primary vs. secondary data, lead to inconsistencies in carbon intensity values.
• These data sources are useful for monitoring trends and assessing decarbonisation progress, but they lack the granularity and transparency needed for policy design and implementation.

Interoperability and Policy Implications

• Ensuring technical and operational interoperability among different standards and methods is challenging but necessary.
• A lack of interoperability could fragment supply chains and increase firms' reporting costs.
• The report highlights the importance of economic principles such as proportionality, innovation, and interoperability to guide the development of carbon intensity metrics.

Conclusion

• Product-level carbon intensity metrics are crucial for designing and implementing targeted mitigation policies, including emissions trading systems, product standards, and carbon border adjustment mechanisms.
• Further development and deployment of clean technologies and improvement of energy mixes are essential for achieving deeper reductions in carbon intensity.
• Interoperable and transparent frameworks are needed to support global trade and effective policy implementation.

Key Figures and Trends

• The steel sector has seen a 10% decline in carbon intensity since 2006, but cement has remained stable.
• Installation-level variation is significant, with EAF and BOF technologies playing a central role in steel production.
• Power inputs and production routes are major drivers of carbon intensity in steel, while the clinker-to-cement ratio is critical in cement.

References

• OECD (2024[3]), (2024[59])
• IFCMA (2024)
• Climate TRACE and CRU data
• US Department of Energy (2022[17])
• WTO (2023[7])

---

Annex and Tables

• Table 1 outlines the main use cases of carbon intensity metrics across governments, households, and firms.
• Annex A provides additional carbon intensity figures for steel and cement, highlighting the impact of weighted vs. unweighted averages and technological differences.