The Impact of Climate Change News on Low-Minus-High Carbon Intensity Portfolios

Abstract

Using a variety of language models and high-quality English-language newspaper
sources, we construct an unexpected climate news index (UCNI) for each model and
source. We measure the impact of these UCNI, plus an Aggregate UCNI over all of
the news sources on a range of Low Carbon Intensity (LCI), High Carbon Intensity
(HCI) and Low-minus-High Carbon Intensity (LmHCI) equity portfolios constructed
by sorting the top 500 US-listed firms by capitalisation (US500) based on their carbon
intensity. We find that the relationship between the UCNI and the LCI, HCI, and
LmHCI portfolio returns is overall not statistically significant for individual news
sources across all of the language models. However, it becomes significant for all
the Aggregate UCNI indices: for the majority of the language models considered the
sensitivity of returns to an increase in the corresponding Aggregate UCNI index, is
negative and statistically significant at the 1% level for HCI portfolio returns but not
significant at the 5% level for LCI portfolio returns over the period from July 2012
to November 2021. We also find that, conditional on the level of the daily UCNI,
the higher the daily UCNI, the higher the average daily annualized return of LmHCI
portfolios. We also generate climate news indices for physical risk and transition risk,
and find that the statistical significance of these indices is close to that of the other
climate change indices for the portfolios analyzed. Overall these results suggest that
the beta of the UCNI factor extracted from an Aggregate news index is a proxy for
a climate risk beta. The UCNI factor can explain the returns of HCI portfolios, and
hence the returns of LmHCI portfolios.

Keywords: Climate-Change

JEL Classification: Q54, G12