Banks made big climate promises but a new study doubts they work

Using European Central Bank lending data, researchers said there was not evidence that voluntary commitments were effective in reducing emissions.

Hundreds of banks, insurers and asset managers vowed to plow $130 trillion in capital into reducing carbon emissions and financing the energy transition as they introduced the Glasgow Financial Alliance for Net Zero. But a recent study, published by the European Central Bank, disputed the effectiveness of those promises.

The working paper by Parinitha (Pari) Sastry, Emil Verner, David Marques-Ibanez, is entitled Business as usual: bank climate commitments, lending, and engagement. The following is a short non-technical  summary that the authors prepared.

Non-technical summary

Financial institutions face increasing pressure from external stakeholders to support the economy’s transition away from carbon-intensive activities. In response banks around the world have joined a range of climate-related initiatives. In this paper, we evaluate the impact of banks’ voluntary climate commitments on their lending behaviour and on the climate impact of borrowing firms. We organize our empirical analysis around three hypotheses for how bank climate commitments can impact their financed emissions. First, climate-aligned banks can divest from polluting firms and reallocate capital to less emission-intensive firms. Second, banks can engage with high-polluting firms to push them to reduce their emissions. A third is that green commitments have limited impact on financed emissions. Our analysis relies on two administrative data sources covering European banks that provide a comprehensive view of these banks’ lending portfolios. The first is a detailed European firm credit registry on bank lending to European firms going back to 2018. The second covers European banks’ global lending by sector.

Our analysis proceeds in three steps. First, we document that there is strong selection into green initiatives. Banks joining green initiatives are larger and lend more to “brown” sectors such as mining. They have a similar share of lending to mining in the euro area as non-signers, but a substantially higher share of lending to mining in their global portfolios. Banks joining the initiative set targets concentrated in power generation, oil and gas, and transport. Moreover, they set targets in sectors to which they have more ex ante lending exposure. Second, We document that climate-aligned lenders reduce lending to targeted sectors by about 20 percent. This would appear to support the hypothesis that lenders divest from brown sectors. Yet we find no evidence of divestment by climate-aligned banks from targeted sectors. We also find no evidence of divestment from other proxies of high-emissions firms, such as firms in the mining sector and firms outside of the EU taxonomy for environmentally sustainable activities. We also show that climate-aligned lenders have a slightly higher rate of entry into new relationships with firms in high-emissions targeted sectors, while we find a limited effect on exit. Further, we find no evidence of a change in interest rates charged by climate-aligned banks to high-emission firms. Across a range of specifications, we find robust evidence against the divestment hypothesis. Third, we find that firms borrowing from NZBA banks are not more likely to themselves set a decarbonization target. Overall, our results cast doubt on the efficacy of voluntary climate commitments for reducing financed emissions, whether through divestment or engagement. This evidence supports recent efforts by governments to improve the credibility of net zero commitments. More broadly, it suggests that voluntary private-sector initiatives may have relatively little impact on decarbonization.

External link