Climate change poses challenges for financial markets and economy. Many policy institutions across the world have recognised these challenges and have been discussing how to update their mandates accordingly. For instance, the ECB (2021) has stated that it is committed to reflect environmental sustainability considerations in its monetary policy. Similarly, the Bank of England (2021) defines its objective as to “play a leading role, through its policies and operations, in ensuring the financial system, the macro-economy, and the Bank of England are resilient to the risks from climate change and supportive of the transition to a net-zero economy” (see also the statements by Brunetti et al. 2021 of the Federal Reserve). These statements indicate that, in near future, the financial regulatory framework will go through an important change to incorporate the climate change, if not it is already happening.
One agreement policymakers have reached about the climate change is that it is a global problem. In line with this agreement, several central banks and supervisors established the Network for Greening the Financial System in 2017 with the purpose of helping to strengthen the global response required to meet the goals of the Paris agreement (Network for Greening the Financial System 2021). Despite this effort, there is still a large heterogeneity across countries in terms of climate policy stringency. In Figure 1, we plot the levels of the Climate Change Performance Index (CCPI) developed by Germanwatch at the country level. The map lays bare the fact that countries have different levels of climate policy stringency. This heterogeneity can be an important factor in the fight against climate change, since it may allow firms to circumvent the higher climate policy stringency in their home country by shifting their operations to less stringent countries. Supporting this concern, Bartram et al. (2021) document that financially constrained firms shifted emissions and output from California to other US states after the introduction of cap-and-trade programme in California.
Figure 1 Country averages of climate policy
Note: Countries with no colour shade are not part of the sample.
In a recent paper (Benincasa et al. 2021), we analyse whether banks exploit this heterogeneity in climate policy stringency across countries. More specifically, we investigate whether banks use cross-border lending to react to higher climate policy stringency in their home countries. From the literature, we know that cross-border lending is an important transmitter of shocks among countries (Cetorelli and Goldberg 2011, Giannetti and Laeven 2012, Ongena et al. 2015, Claessens 2017, Hale et al. 2020, Doerr and Schaz 2021). Therefore, banks may increase their cross-border lending when faced with more stringent climate policy in their home country.
We find that banks do indeed react to greater climate policy stringency in their home country by increasing their cross-border lending. As an illustration of our findings, Figure 2 shows a strong positive relationship between cross-border loan ratios on bank balance sheets and home country climate policy stringency. Our regression analysis suggests a large effect: banks increase their shares in cross-border syndicated loans by 10% if policy of their home country increases by same level experienced in the US between 2007 and 2017. Overall, our results depict a clear picture in which banks use cross-border lending as a regulatory arbitrage tool against climate policies, which may reduce the effectiveness of such policies.
Figure 2 Home country climate policy and cross-border bank lending
As indicated before, our measure of climate policy stringency is the CCPI (developed by Germanwatch e.V. with the aim of tracking efforts to combat climate change in 57 countries and the EU). As argued by Delis et al. (2019), a measure for the stringency of country climate policy should account for both the ambition and the effort of the government policy itself. The former is measured by the efficiency of the policy, while the latter is measured by the effectiveness of the policy in reaching specific outcomes. Therefore, providing a complete picture of countries' climate protection action efforts, the CCPI has been employed by other studies to measure countries' climate policy stringency (Delis et al. 2019, Atanasova and Schwartz 2019, Lin et al. 2020). The CCPI comes with two main advantages. First, being a weighted average of 14 different climate policy indicators, the CCPI is a broad and inclusive assessment of the countries’ climate policy stringency. Second, it facilitates climate policy comparison of countries with different backgrounds as it summarises the differences with one metric. We combine the CCPI with syndicated loan data, which we use to assess bank cross-border lending. Syndicated loans are one of the main tools for cross-border lending (De Haas and Van Horen 2013). In addition, syndicated loans make cross-border lending easier for smaller banks too, as the lead arranger of a syndicated loan can take actions to reduce the information asymmetries. Therefore, a combination of the CCPI series and syndicated loan data provides us with a relevant setting to investigate whether banks alter their cross-border lending to react to a change in climate policy stringency.
A regression model in which shares in cross-border loans are regressed on CCPI can account for the effects of loan demand and other country-level variables, in addition to the effect of CCPI. For instance, observing an increase in the CCPI of a country, a firm may increase its loan demand to the banks from that country. This can occur as the firm may want to use the relationship with a bank from a high CCPI country as a signalling device. Similarly, this increase in loan demand can be driven by firm’s desire to increase its knowledge in efforts against climate change and a lending relationship with this bank can provide this knowledge. These arguments imply that without properly controlling for loan demand, the relationship between the CCPI and cross-border lending cannot be interpreted in terms of the loan supply. In our preferred specification, we control for all loan demand driven aspects, such as borrowers’ and loans’ characteristics, and identify an effect that reflects banks’ loan supply.
In addition to the loan demand, country-level characteristics that are correlated with both climate policy stringency and cross-border lending can create a bias in our estimations. For instance, an improvement in economic conditions can lead to an increase in both the CCPI and cross-border lending, or a change in demographics of the country can affect the CCPI by altering the perception of the climate change and cross-border lending by affecting loan demand. To mitigate such concerns, we collect information about country level economic conditions, culture, legal environment, and demographics and include these variables into our models. Our results do not change when we control for these variables.
We find evidence supportive of the aforementioned mechanism – i.e. the increase in cross-border lending is driven by a regulatory arbitrage by exploiting the heterogeneity among the lenders. First, we document that the positive effect of climate policy stringency on cross-border lending occurs only if the home country of the lender has a more stringent climate policy compared with the borrower's country. This finding indicates that the banks use the cross-border lending as a device to mitigate the effects of the climate policy since it shows that banks increase their cross-border lending selectively. Second, we find that banks that are expected to engage with cross-border lending as a reaction to climate policy stringency are indeed the ones who are more likely to do so. For instance, the magnitude of the effect is significantly larger for the banks that have higher cross-border loans in their books and for banks that face a higher non-performing loans ratio (NPL). A higher cross-border loans ratio implies that the bank has more experience with cross-border lending, which means that it is easier for this bank to use cross-border lending to react to changes in domestic climate policy stringency. A higher NPL ratio creates a stronger incentive for the bank to engage with cross-border lending since more stringent climate policy can reduce the returns of the loans when the bank needs a higher return rate due to the high NPL ratio.
We continue our analysis by examining which category of the CCPI is more important for the cross-border lending. The CCPI has four categories: greenhouse gas emissions, renewable energy, energy use, and climate policy. Estimating horse-race regression models that includes these four categories, we find that climate policy is the most important category for cross-border lending. This implies that banks react to the actual measures taken by the respective domestic governments, instead of the realised outcomes of these measures in terms of emissions for example. This finding also lends support to our interpretation that the underlying mechanism of our findings is capturing regulatory arbitrage.
Overall, our paper provides two main new insights. First, banks may be taking actions to reduce the influence of climate policy stringency on their loan portfolio. Second, the heterogeneity in climate policy stringency among countries can induce cross-border lending due to regulatory arbitrage opportunities it creates, which indicates that lack of homogeneity in the regulations for climate change can reduce the effectiveness of such regulations through a bank lending channel.
Atanasova, C and E S Schwartz(2019), “Stranded fossil fuel reserves and firm value,” NBERWorking Paper No. 26497.
Bank of England (2021), "Climate change".
Bartram, S M, K Hou, and S Kim (2021), “Real effects of climate policy: Financial constraints and spillovers,” Journal of Financial Economics.
Benincasa, E, G Kabaș, and S Ongena (2021), “There is no planet B”, but for banks there are “countries B to Z”: Domestic climate policy and cross-border bank lending", CEPR Discussion Paper 16665.
Brunetti, C, B Dennis, D Gates, D Hancock, D Ignell, E K Kiser, G Kotta, A Kovner, R J Rosen and N K Tabor (2021), “Climate Change and Financial Stability”, FEDS Notes, 19 March.
Cetorelli, N and L S Goldberg (2011), “Global banks and international shock transmission: Evidence from the crisis”, IMF Economic Review, 59(1): 41-76.
Claessens, S (2017), “Global banking: Recent developments and insights from research,” Review of Finance, 21(4): 1513-1555.
De Haas, R and N Van Horen (2013), “Running for the exit? International bank lending during a financial crisis,” The Review of Financial Studies, 26(1), 244-285.
Delis, M D, de K Greiff and S Ongena (2019), “Being stranded with fossil fuel reserves? Climate policy risk and the pricing of bank loans”, EBRD Working Paper No. 231.
Doerr, S and P Schaz (2021), “Geographic diversification and bank lending during crises,” Journal of Financial Economics, 140(3): 768-788.
ECB (2021), “ECB presents action plan to include climate change considerations in its monetary policy strategy”, press release, 8 July.
Giannetti, M and L Laeven (2012), “The flight home effect: Evidence from the syndicated loan market during financial crises,” Journal of Financial Economics, 104(1): 23-43.
Hale, G, Kapan and C Minoiu (2020), “Shock transmission through cross-border bank lending: Credit and real effects,” The Review of Financial Studies, 33(10): 4839-4882.
Karolyi, G A and A G Taboada (2015), “Regulatory arbitrage and cross‐border bank acquisitions,” The Journal of Finance, 70(6): 2395-2450.
Lin, C, T Schmid and M S Weisbach (2020), “Climate Change and Corporate Investments: Evidence from Planned Power Plants”, Fisher College of Business Working Paper, (2019-03), 026.
Network for Greening the Financial System (2021), "Origin and purpose".
Ongena, S, J L Peydró and N Van Horen (2015), “Shocks abroad, pain at home? Bank-firm-level evidence on the international transmission of financial shocks,” IMF Economic Review, 63(4): 698-750.
Ongena, S, A Popov and G F Udell (2013), ““When the cat's away the mice will play”: Does regulation at home affect bank risk-taking abroad?” Journal of Financial Economics, 108(3): 727-750.