The pricing of carbon transition risk is a key question as investors consider climate-aware investments. Accurate risk pricing can support climate change mitigation, potentially reducing the need for heavy-handed government intervention. In theory, brown firms are more exposed to the transition and policy risk and should earn higher expected returns in equilibrium. However, green firms can outperform in the transition to the net-zero economy when policy shocks kick in, consumer attention turns, and investor tastes shift. Alternatively, if investors do not materially pay attention to carbon footprint, we would not observe significant outperformance by either green or brown firms. Despite substantial interest, both academics and practitioners continue to debate the extent to which financial markets have priced in carbon transition risk.

In this paper, I revisit the carbon return—the return spread between brown and green firms. The measure of carbon transition risk is the carbon intensity or emissions per unit of sales, and emissions data are only available to investors with significant lags. After accounting for the data release lag, the carbon return is significantly negative in the U.S. It varies considerably across countries following climate preference shifts and variations in climate policy tightness. Overall, the evidence suggests that the transition to full carbon-aware pricing is still early and underway.

Measuring Carbon Transition Risk

I measure carbon transition risk with the carbon intensity or emissions per unit of sales. The intensity measure is superior to total emissions because total emissions equal a weighted sum of sales scaled by emission factors and grow almost one-for-one with firm sales in data. Instead, the carbon intensity distinguishes firms’ environmental impact from firm size and operations.

A key empirical challenge in constructing the measure is the real-time measurement of emissions known to investors due to the gradual release and revisions of the carbon data. Since emissions contain substantial information about firm performance, they should be lagged sufficiently to avoid the forward-looking bias. I provide the first empirical assessment of the carbon data release lags based on the S&P Trucost and find that the emissions data are only available to investors with significant lags. The median lags are 10 and 12 months after the emission fiscal year-end for the U.S. and international samples, respectively.

Carbon Returns in the U.S. and Internationally

 Exhibit 1: U.S. Cumulative Excess Carbon Return

Using the most recent carbon emission data available to investors, I show that more carbon-intensive firms earned significantly lower returns than less carbon-intensive ones in the U.S. from 2009 to 2021. The monthly value-weighted carbon return spreads are -0.39% and -0.27% for scope 1 and 2 carbon intensities. Exhibit 1 shows that a trading strategy that longs more carbon-intensive firms and shorts less carbon-intensive ones loses more than half of its initial value over the sample period. The negative carbon return is robust to factor adjustments and various robustness checks. Globally, brown firms again tend to underperform, though the carbon return is insignificant.

It is noteworthy that my findings differ from previous studies because I relate stock returns to lagged, rather than contemporaneous, carbon footprints. Prior studies suggest that Brown firms appear to outperform because firms with high sales growth have higher total emissions, resulting in a “browner” profile and stronger contemporaneous stock performance. This seeming outperformance of brown firms is due to the use of forward-looking data and should not be interpreted as evidence of genuine investor commitment to climate-aware investments.

Exhibit 2: Country-Level Abnormal Mean Carbon Returns

Next, I delve into country-level carbon returns to examine whether markets have priced in carbon transition risk. Exhibit 2 shows that the carbon return exhibits a huge dispersion across countries and is lower in developed markets than in emerging markets. The international carbon returns can reflect variations in expected risk premia but can also reflect various unanticipated in-sample shocks. Specifically, developed countries have experienced stronger growth in climate concerns, measured by country-level sustainable flows and surveyed climate concerns, generating lower carbon returns in these countries. After controlling for all in-sample shocks, carbon returns tend to be higher in countries with climate policy tightness, reflecting compensation for heightened policy risk as in equilibrium. Overall, the evidence suggests that investors have started pricing in carbon transition risk, but the carbon premium can be muted for an extended period as the carbon transition takes place.

In summary, despite recent progress, financial markets have not priced in carbon transition risk as effectively as one might hope. Throughout the transition, carbon returns can fluctuate significantly. Future returns not only depend on the expected carbon premium in equilibrium but also vary with future shifts of public attention to climate change, investor preference, and climate policy. It is, therefore, crucial to communicate to investors the properties of green returns during the transition and to caution against forming expectations about future carbon returns based on past performance.

From a policy point of view, financial markets alone are unlikely to drive the scale of climate action required. The burden of combating climate change rests heavily on policymakers. Highly restrictive policies, such as strict carbon quotas, may fail to produce real carbon reductions, as firms could avoid these constraints by spinning off operations. In contrast, policies that align with consumer sentiment shifts, such as carbon taxes or cap-and-trade systems, may provide a more effective path to achieving climate goals.