Chimneys emitting smoke and financial charts
VoxEU Column Environment Financial Markets

Why carbon emissions are associated with higher stock returns

Carbon transition risk must be priced in financial markets for a shift to a low-carbon economy. This column explores the relationship between carbon emissions and earnings surprises to help understand the source of the carbon premium. Carbon emissions have a similar association with earnings surprises as they do with stock returns. Both the level of and change in emissions are positively related to earnings surprises. The market may not be fully pricing in carbon transition risk, and government intervention may be critical to achieving the carbon transition.

Whether carbon transition risk is priced in financial markets has first-order implications for the likelihood and speed of a shift to a low-carbon economy. If stocks exposed to transition risk (as proxied by high carbon emissions) are heavily discounted, then companies have strong incentives to cut their emissions, investors will engage with companies to lower their emissions, and emitters will have difficulty raising capital. In contrast, if markets insufficiently price in transition risk, then companies may not reduce their emissions nor investors decarbonise their portfolios.

An influential paper by Bolton and Kacperczyk (2021) finds that US companies with high levels of and changes in carbon emissions have high realised stock returns. These results are consistent with such firms facing a high cost of equity and thus markets pricing in transition risk. However, Aswani et al. (2024a) show that this carbon premium becomes insignificant when either studying carbon intensities (emissions scaled by sales), or focusing on disclosed rather than estimated emissions. Bolton and Kacperczyk (2024) respond that absolute emissions are the relevant measure of transition risk; Aswani et al. (2024b) disagree.

Regardless of whether level, change, or intensity is the appropriate measure and whether estimated emissions are reliable, these results assume that realised returns are a good proxy for expected returns and thus the cost of capital. In addition to the authors themselves, 18 papers published in the Journal of Finance, Journal of Financial Economics, Review of Financial Studies, Review of Finance, Journal of Financial and Quantitative Analysis, Management Science, and Annual Review of Financial Economics since 2020 refer to Bolton and Kacperczyk’s results as documenting higher ‘expected returns’, a ‘risk premium’, ‘carbon risk[s]’, ‘climate risk[s]’, or that ‘risk is priced’.

The interpretation is surprising since advocates of environmental, social, and governance (ESG) investing typically use the high returns to certain green companies as evidence that ESG pays off – that ESG is good for firm value and underpriced by the market, rather than bad for firm value and exposing companies to excessive risk. Indeed, a large literature on ESG investing uses realised abnormal returns as a measure of unexpected returns and thus outperformance rather than risk. For example, Gompers et al. (2003) document high returns to well-governed companies, Fornell et al. (2006, 2016) to firms with high customer satisfaction, Edmans (2011, 2012) and Boustanifar and Kang (2022) to stocks with high employee satisfaction, and Lins et al. (2017) to high-trust businesses in the financial crisis. The interpretation is also surprising since it is theoretically unclear whether sustainability will affect the cost of capital at all; the main effects are likely to be on cash flows (Edmans 2021).

A standard way to disentangle outperformance from risk is to study future earnings surprises. La Porta et al. (1997) find that value companies systematically beat analyst expectations. In an ESG context, Core et al. (2006) show that well-governed firms do not deliver positive earnings surprises; Giroud and Mueller (2011) find that they do in non-competitive industries. Edmans (2011) documents positive earnings surprises for companies with high employee satisfaction, Edmans (2023) finds similar results in non-US countries with flexible labour markets, and Fornell et al. (2016) uncover analogous findings for stocks with high customer satisfaction.

In a new paper (Atilgan et al. 2024), we study the relationship between carbon emissions and earnings surprises to help understand the source of the carbon premium. We find that carbon emissions have a remarkably similar association with earnings surprises as they do with stock returns. Both the level of and change in emissions are positively related to earnings surprises, just as Bolton and Kacperczyk find with realised returns. An increase of one standard deviation in the level of scope 1, scope 2, or scope 3 emissions is associated with an increase in the one-year earnings surprise that is approximately twice its sample median and significant at the 1% level. In contrast, carbon intensities are unrelated to earnings surprises, as are emissions levels and changes when focusing on disclosed emissions only, consistent with Aswani et al. (2024a). These results suggest that the carbon premium, where it exists, at least partly results from outperformance.

We find similar results when relating carbon emissions to three-day earnings announcement returns. An increase of one standard deviation in scope 1 emissions levels (changes) is associated with a 0.12% (0.19%) higher announcement return, controlling for size and book-to-market effects. Analogous figures are 0.20% (0.23%) for scope 2 emission levels (changes) and 0.45% (0.43%) for scope 3 emission levels (changes). The four quarterly earnings announcements per year account for 30–50% of the carbon premium based on both levels and changes.

Our results suggest that the market is not fully pricing in carbon transition risk, casting doubt on whether market forces alone can bring about the shift to a low-carbon economy. This may be because companies and investors view carbon emissions as an externality that harms society but not the polluting firms, even in the long term. Thus, some firms choose not to invest in lowering their emissions and enjoy higher earnings and stock returns as a result. While it is frequently claimed that ‘climate risk is investment risk’, the risk to society may not be fully borne by investors. These findings highlight the criticality of government intervention to achieve the carbon transition, and the trade-off that investors face between fiduciary duty and net-zero alignment in the absence of such action (see also Gosling and MacNeil 2023).

Beyond the topic of carbon emissions, our research has implications for research on ESG more generally. While the practice of ESG has enjoyed tremendous growth, it has suffered a recent backlash, in part due to the opportunistic interpretation of evidence. If a study finds high returns to ‘green’ companies, some ESG advocates interpret it as evidence that being green pays off. But if the study finds high returns to ‘brown’ companies, they would interpret it as evidence that being brown leads to a higher cost of capital. This is an example of confirmation bias, where people interpret results however they want to and ignore alternative explanations (see Edmans 2024 for a further analysis of the problem and how to address it). Instead of rushing to our preferred explanation of the data, the scientific process involves taking all potential interpretations seriously.


Aswani, J, A Raghunandan, and S Rajgopal (2024a), “Are carbon emissions associated with stock returns?”, Review of Finance 28: 75–106.

Aswani, J, A Raghunandan, and S Rajgopal (2024b), “Are carbon emissions associated with stock returns? Reply”, Review of Finance 28: 111–15.

Atilgan, Y, A Edmans, K O Demirtas, and A D Gunaydin (2024), “Does the carbon premium reflect risk or mispricing?”, working paper.

Bolton, P, and M Kacperczyk (2021), “Do investors care about carbon risk?”, Journal of Financial Economics 142: 517–49.

Bolton, P, and M Kacperczyk (2024), “Are carbon emissions associated with stock returns? Comment”, Review of Finance 28: 107–9.

Boustanifar, H, and Y D Kang (2022), “Employee satisfaction and long-run stock returns: 1984–2020”, Financial Analysts Journal 78: 129–51.

Core, J E, W R Guay, and T O Rusticus (2006), “Does weak governance cause weak stock returns? An examination of firm operating performance and investors’ expectations”, Journal of Finance 61: 655–87.

Edmans, A (2011), “Does the stock market fully value intangibles? Employee satisfaction and equity prices”, Journal of Financial Economics 101: 621–40.

Edmans, A (2012), “The link between job satisfaction and firm value, with implications for corporate social responsibility”, Academy of Management Perspectives 26: 1–19.

Edmans, A (2021), “Does sustainability reduce the cost of capital”,, 17 March.

Edmans, A (2023), “Applying economics – not gut feel – to ESG”, Financial Analysts Journal, forthcoming.

Edmans, A (2024), May contain lies: How stories, statistics, and studies exploit our biases – and what we can do about it, Penguin Random House

Fornell, C, S Mithas, F V Morgeson III, and M S Krishnan (2006), “Customer satisfaction and stock prices: high returns, low risk”, Journal of Marketing 70: 3–14.

Fornell, C, F V Morgeson III, and G T M Hult (2016), “Stock returns on customer satisfaction do beat the market: Gauging the effect of a marketing intangible”, Journal of Marketing 80: 92–107.

Giroud, X, and H M Mueller (2011), “Corporate governance, product market competition, and equity prices”, Journal of Finance 66: 563–600.

Gompers, P, J Ishii, and A Metrick (2003), “Corporate governance and equity prices”, Quarterly Journal of Economics 118: 107–56.

Gosling, T, and I MacNeil (2023), “Can investors save the planet? – NZAMI and fiduciary duty”, Capital Markets Law Journal 18: 172–93.

La Porta, R, J Lakonishok, A Shleifer, and R Vishny (1997), “Good news for value stocks: Further evidence on market efficiency”, Journal of Finance 52: 859–74.

Lins, K V, H Servaes, and A Tamayo (2017), “Social capital, trust, and firm performance: The value of corporate social responsibility during the financial crisis”, Journal of Finance 72: 1785–824.