Heterogeneity across different jurisdictions in climate policies to regulate firms’ emissions can lead to economic activity shifting away from countries with the most stringent environmental laws. This column uses the case of the US, where California’s cap-and-trade programme is the only policy regulating emissions by industrial firms, to show that carbon leakage can occur through the rewiring of supply chains. Uncoordinated climate action can increase the probability of relationships with suppliers subject to climate policies being terminated (or not being initiated). Ultimately, it can alter the structure of production networks and affect the regulation of scope 3 carbon emissions.
Carbon leakage occurs if businesses move production from countries with strict climate policies to others with laxer ones. 1 This phenomenon has become of primary concern for regulators who are facing a set of geographically bounded climate policies targeting the emissions of firms that operate at a global scale through their supply chain networks (see, for example, the efforts to address carbon leakage within the EU Emissions Trading System).
Corroborating these concerns, previous research has provided evidence of practices leading to carbon leakage. For example, some firms relocate production and their economic activity and scope 1 carbon emissions internally, from facilities in jurisdictions with regulated (higher) carbon prices to those in jurisdictions with unregulated (lower) carbon prices (Ben-David et al. 2021, Benincasa et al. 2021, Bartram et al. 2022, Laeven and Popov 2023, Ivanov et al. 2024).
Our study
In a recent paper (Benincasa et al. 2024), we uncover an additional economic mechanism leading to carbon leakage: the rewiring of supply chains away from carbon-pricing regimes. Specifically, we study uncoordinated climate action in the US, where the California cap-and-trade programme is the only policy that regulates emissions produced by industrial firms. In this context, we show that customers are more likely to terminate their supply chain relationships with firms regulated by the programme than with competing suppliers operating outside the regulated area.
What are the possible determinants of the rewiring of supply chains away from climate policies? Two opposing forces may determine such an outcome. On the one hand, uncoordinated climate action exposes suppliers to different degrees of climate transition risks. As firms subject to climate policies suffer erosion in profitability and increase in costs (Ryan 2012), customers may be incentivised to shift their inputs sourcing to competing suppliers. 2 On the other hand, this incentive can fade away if regulators grant climate policy exemptions to highly energy-intensive and trade-exposed sectors (Fowlie et al. 2016); if supplier switching costs are high (Barrot and Sauvagnat 2016, Bernard et al. 2019); or if environmentally conscious stakeholders push customers to keep ties with suppliers subject to climate policies to ultimately decarbonise their supply chains (Krueger et al. 2020, Aghion et al. 2023, Homroy and Rauf 2024).
Studying supply chains in a context of uncoordinated within-country climate action
To study the net effect of carbon policies on supply chains as a result of these two opposing forces, we conduct an empirical study of suppliers subject to the California cap-and-trade programme implemented in 2013. This programme provides us with an ideal setting of uncoordinated within-country climate policies, since no other US state has a comparable programme in place to regulate the emissions of industrial firms. The cap-and-trade programme sets a cap on emissions of electricity generation and industrial facilities that produce more than 25,000 tonnes of carbon dioxide equivalent at their California facilities. In 2015, the programme was expanded to include petroleum and natural gas distributors. These firms need to cover their excess emissions by purchasing permits to emit.
In our empirical approach, we compare the likelihood that customers sourcing from suppliers subject to the cap-and-trade programme terminate their business relationship with that of otherwise similar suppliers unaffected by the policy. We ensure the comparability of suppliers subject to the programme and other suppliers along several dimensions such as industry, firm size, and profitability. Figure 1 illustrates our empirical strategy.
Figure 1 Illustration of our empirical methodology
Supply chains rewire away from suppliers subject to the California cap-and-trade
Our results, summarised in Figure 2, indicate a 5 percentage point higher likelihood that clients terminate their pre-existing customer relationships with suppliers subject to the cap-and-trade programme, compared to otherwise similar suppliers outside of the programme. This is an economically sizeable effect, as only about 20% of the business relationships are ended each year for the average supply chain. We also document that California suppliers with large emissions relative to their assets – and thus more profoundly affected by the cap-and-trade programme due to the likely higher expenditures allocated to emission permits – are proportionally more affected by business relationship terminations. We interpret these results as a rewiring of supply chains away from the programme, and we further corroborate this interpretation of our results, showing that suppliers subject to the programme are also 8 percentage points less likely to initiate new customer relationships relative to otherwise similar suppliers not affected by the policy.
Figure 2 Probability of termination of supply chain relationships around the introduction of the cap-and-trade programme
Substitutable suppliers and suppliers with less environmentally attentive customers are mostly affected
We uncover significant variation in the effect across different suppliers. For example, we find that highly substitutable suppliers (such as those operating in highly competitive industries, or in sectors producing standardised inputs) exhibit a greater probability of suffering the termination of supply chain relationships. This is consistent with customers’ incentives to rewire supply chains being stronger when switching costs are lower.
Our evidence suggests that the uncoordinated climate action within the US places suppliers subject to the programme at a disadvantage relative to their competitors. Specifically, we find that affected suppliers experience larger declines in revenues, assets, and profitability, along with a greater increase in the average costs of goods sold, than comparable non-affected suppliers. 3
Our results also show that the most environmentally aware customers – such as those with a high index of climate change attention, as proposed by Sautner et al. and Zhang (2023), or those headquartered in Democrat-led US states – are not as likely to rewire their supply chains away from the cap-and-trade programme.
Uncoordinated climate action produces unintended consequences for suppliers’ emissions and the supply chain emissions of their customers
Finally, we shed light on the environmental implications of uncoordinated climate action in the US. We find that the overall emission intensity of suppliers subject to the programme does not decrease over the three-year period following the introduction of the policy; if anything, emission intensity increases. However, the upstream supply chain emission intensity of customers indirectly affected by the policy (i.e. the upstream supply chain emission intensity of clients sourcing from affected California suppliers) increases following the introduction of the policy.
Conclusion
Our paper shows that uncoordinated climate action in the US leads to carbon leakage through supply chain rewiring. Customers move away from suppliers subject to California’s cap-and-trade programme, and new clients are less likely to establish new business relationships with affected suppliers. The rewiring of supply chains induced by climate policies is mainly concentrated within business relationships characterised by highly substitutable suppliers and less climate-attentive customers. Finally, uncoordinated within-country climate action may have adverse environmental consequences, such as an increase in the upstream scope 3 emissions intensity of customers indirectly exposed to the policy. Overall, our research contributes to the ongoing policy discussion on the need to coordinate climate action to improve the effectiveness of climate policies. In other carbon pricing regimes, such as the EU Emissions Trading System, these discussions resulted in the introduction of a carbon border adjustment mechanism to address carbon leakage concerns. In the context of the California cap-and-trade policy, a first step towards the prevention of carbon leakage might be the extension of the carbon market across the entire country.
Source : VOXeu
