TY - JOUR T1 - Sustainability Attribution: The Case of Carbon Intensity JF - The Journal of Impact and ESG Investing SP - 93 LP - 99 DO - 10.3905/jesg.2021.1.027 VL - 2 IS - 1 AU - Guido Bolliger AU - Dries Cornilly Y1 - 2021/08/31 UR - https://pm-research.com/content/2/1/93.abstract N2 - This article proposes a method to decompose the carbon intensity of a portfolio with respect to a benchmark into an allocation and a selection component. The carbon-intensity decomposition allows for a better understanding of the sources of the difference between the carbon footprint of a portfolio and that of its benchmark. As such, it prevents greenwashing by analyzing whether the carbon exposure of a portfolio results from active stock selection choices on the part of the manager or from passive sector exclusion decisions. The authors’ approach is based on methods developed for traditional performance attribution. They discuss an equity example using the MSCI ACWI Sustainable Impact Index and a fixed-income example around the ICE BofA Global Corporate Green Bond Index. In the latter example, they show that a higher portfolio carbon intensity does not necessarily contradict the portfolio’s stated environmental, social, and governance (ESG) or impact objectives. Their methodology can easily be extended to any other sustainability or impact metric that is constructed as a weighted average of asset scores, thus providing greater precision in analyzing the sources and implications of incorporating ESG into portfolio construction.TOPICS: ESG investing, portfolio construction, security analysis and valuation, fixed income and structured financeKey Findings▪ The difference in carbon intensity between a portfolio and its benchmark can be decomposed into an allocation and a selection component.▪ The higher intensity of impact investment strategies stems mostly from the allocation component. The selection component shows that impact portfolios fulfill their mandates. ▪ The attribution methodology can be applied to any portfolio score that is constructed as a weighted average of asset scores (e.g., ESG scores). ER -