Female representation on boards and carbon emissions: International evidence

What this research asks

Does the make-up of a company’s board show up in its environmental record? Specifically: do firms with more women on their boards emit less carbon relative to the size of their business?

Boards approve strategy, capital spending and risk appetite, which are the decisions that ultimately determine how carbon-intensive a company is. If who sits on the board changes how those decisions get made, it should be visible in the emissions data. This paper tests that across a large, international sample.

How it was studied

• Sample: 10,844 firm-year observations from 1,514 carbon-reporting companies in 43 countries, 2005–2019.
• Emissions: direct (Scope 1) and indirect energy (Scope 2) emissions, measured as the log of total CO₂-equivalent tonnes divided by revenue. That is emission intensity, not absolute emissions. Source: Thomson Reuters ASSET4.
• Board gender diversity: the proportion of female directors, with two alternatives tested for robustness: whether a board has at least one woman, and the number of women on the board. Source: Refinitiv Datastream.
• Controls: firm size, profitability, leverage, capital expenditure, asset tangibility, R&D intensity, earnings volatility, Tobin’s Q and market-to-book, plus country-level GDP per capita and progress towards UN Sustainable Development Goal 13 (climate action). Year, industry and country fixed effects throughout.

What we found

1. Firms with a higher proportion of women on the board have lower carbon emission intensity. In the baseline model the relationship is negative and statistically significant at the 1% level.
2. The result holds across four estimation approaches: pooled OLS with firm-clustered standard errors, Fama–MacBeth, Newey–West, and weighted least squares.
3. It survives the obvious objections. It holds using the alternative diversity measures, after adding other board characteristics (board size, independence and CEO duality), using industry-adjusted variables, and after removing the United States and Japan, which together make up almost half of the sample.
4. It holds on a matched sample. Using propensity score matching to compare firms with above- and below-median board gender diversity that are otherwise similar, the relationship remains negative and significant at the 5% level. That makes omitted-variable bias less likely to explain the result, though, as with any observational study, it cannot rule it out entirely.
5. Other firm characteristics matter in expected ways. Larger, more leveraged, more capital-intensive firms and firms with more tangible assets emit more per unit of revenue. More profitable firms and firms that spend more on R&D emit less.

Why it matters

For investors and policymakers, the paper adds cross-country evidence to the argument that board composition is a material governance variable, not just a disclosure item. It connects to live debates on board diversity requirements and on how investors assess carbon exposure.

@article{rjiba2022female,
  title   = {Female representation on boards and carbon emissions: International evidence},
  author  = {Rjiba, Hatem and Thavaharan, Tharshan},
  journal = {Finance Research Letters},
  volume  = {49},
  pages   = {103079},
  year    = {2022},
  doi     = {10.1016/j.frl.2022.103079}
}