Green ECB Credit: One Step Too Far

Refinancing of green credit by the European Central Bank at favourable conditions has received the backing of some economists and policymakers. The article examines the motives put forward to support this proposal, as well as how it could be implemented and how its proponents try to ward off potential criticisms. A number of the issues raised by the proposal are then discussed from a logical, economic and a legal point of view. The article concludes that adopting the proposed measure would be going one step too far.

There have been recent calls to “green” monetary policy through the refinancing of “green” firms at privileged conditions by the European Central Bank (Colesanti Senni et al., 2023, 2024; Kedward et al., 2022a, 2022b; Monnet & van ’t Klooster, 2023). In France, this proposal has been supported at the head of state level (Macron, 2023) and echoed in columns in the daily newspaper Le Monde (van ’t Klooster & Monnet, 2023; Collectif, 2024). In the following, I refer to this proposal as “the proposed measure”.

The proposal

Motives

The motives that are put forward to justify the proposed measure vary from one publication to the other. However, most, but not all, motives relate to the conduct of monetary policy.

Monetary policy motives

From a historical point of view, Colesanti Senni et al. (2023) point out that the targeted longer-term refinancing operations (TLTROs) of the ECB have been successful in supporting the distribution of credit but that the carbon emission intensity of the second tranche of the TLTRO III programme was high. The instrument would thus be intrinsically good but its design would have to be amended to take climate change into account. However, the authors identify green credit as credit distributed to firms that belong to green sectors and brown credit is distributed to firms in brown sectors. This is an oversimplification that I elaborate on in the second part of this article. Monnet and van ’t Klooster (2023, p. 6) also mention targeted credit policy that favoured exports in Europe in the past. However, they neither evaluate these policies, nor explain why they were in part implemented by central banks, with investment banks or government agencies also contributing in some cases, nor why they were discontinued.

From a geographic point of view, the literature supporting the proposed measures points to various programmes, mostly in Asian emerging economies (China, India, Malaysia, Singapore, South Korea), but also in Japan, involving central bank refinancing at privileged rates under the condition that credit is granted to green projects. However, in most of these economies, the institutional settings differ from those in Europe, with less central bank independence, and less developed and more tightly administered financial markets – differences the proponents of the measure do not mention.¹

From a topical point of view, the need to protect green credit is often related to the high level of interest rates (Colesanti Senni et al., 2024; Monnet & van ’t Klooster, 2023). However, withdrawing the proposed measure when interest rates fall is not suggested.

From a more structural point of view, Monnet and van ’t Klooster (2023) argue, in a slightly paternalistic tone, that “a lack of sufficient green investment is undesirable in terms of ensuring long-term price stability” (p. 1). This would call for a “disinflationary credit policy” (Monnet & van ’t Klooster 2023, p. 4), as embodied in the proposed measure. However, this belongs to a non-monetary approach of inflation discussed in the second part of this article. Monnet and van ’t Klooster (2023) also insist that credit policy would be needed because “undifferentiated interest rates are far from neutral” (p. 4). However, neutrality is a myth if it is understood as taking decisions that have the same or very comparable effects on all or most of economic agents and projects, as is seemingly implied by the authors. What central banks refer to regarding neutrality is rather making decisions that do not deliberately aim at having distributive impacts: the reference is thus taken from market allocation, rather than a supposedly ideal allocation by a social planner.

Other motives

Another motive relates to financial stability considerations. For instance, Colesanti Senni et al. (2023) state that “[b]y introducing a green TLTRO, the ECB can incentivize banks to lend to green firms and thus induce a transformation in the economy. This ultimately reduces financial stability risks.” However, in contrast to Pfister and Valla (2021), the authors do not explain why such an incentive could not be provided more directly and efficiently by prudential policies.

A more or less explicit motive is that of a market failure. Kedward et al. (2022b) suggest to not just rely on price incentives, but also to direct credit creation in favour of sectors dictated by green indusrial strategy, banning any distribution of credit for dirty sectors, defined by a public taxonomy. Again, this approach is in stark contrast with Pfister and Valla (2021), who instead suggest that economic agents should be provided with the right incentives, thus allowing financial markets to play a useful role in allocating saving in line with social welfare.

One last sort of motive is of a legal nature: the ECB would have no choice because, according to Article 127(1) of the Treaty on the Functioning of the European Union (TFEU), “Without prejudice to the objective of price stability, the ESCB shall support the general economic policies in the Union”.² Among these policies are those aiming at the improvement of the quality of the environment. This point is discussed below.

Implementation

In the simplest form, as in Colesanti Senni et al. (2023), the ECB would conduct green TLTROs, imposing differentiated interest rates depending on the emission intensity of banks’ loans.

Monnet and van ’t Klooster (2023) propose a more complex arrangement which could include: a credit line provided by the central bank at a concessional rate to a government agency or investment bank distributing green loans accepted as collateral; green refinancing credit at a long maturity (thus very similar to green TLTROs, although the maturity could seemingly be even longer); green asset purchase programmes and green reserve requirements, with differentiated deposit rates based on the volume of green lending by the bank.³

Finally, Kedward et al. (2022a, 2022b) propose a grandiose construction, which would apply to both banks and what they call “institutional capital” (i.e. insurance companies, funds, etc.). It would combine indirect price-based policies, including sector-targeted refinancing lines by the central bank; direct price-based policies, such as interest rate floors and ceilings or subsidised credit to households, SMEs or priority sectors; and direct quantity-based policies, such as portfolio restrictions, credit quotas, large scale public investment through investment banks. In line with their lack of confidence in market adjustments, the authors express a clear preference for direct policies.

Dismissal of potential criticisms

The supporters of a green ECB credit try to present their proposal as compatible with both democracy and the independence of the central bank.

Democracy

A debatable reference to “democracy” is often put forward as a justification of the proposed measure. However, this measure would delegate a distributive role to the central bank, which is not an elected body, at the expense of parliaments and governments, as shown below, with the attending risk of politicising its decision-making process. Monnet and van ’t Klooster (2023) try to solve this issue through the creation of credit councils.⁴ Kedward et al. (2022b) also suggest that “to ensure the effective design and coordination of green credit policy, new national public agencies comprised of representatives from central banks and relevant financial supervisory bodies and ministries of finance, industry and environment/climate may be needed” (p. 8). In both instances, the body in charge of implementing the policy would basically be an unelected administrative committee, even if some members are parliamentarians, as was the case with the French Conseil National du Crédit (see footnote 4). This body would mimic the behaviour of a parliament or government, without having the same legitimacy.

Independence of the central bank

Rather euphemistically, Colesanti Senni et al. (2023) state that “since ample existing policies exist on the EU-level concerning the overall orientation of climate policy, there is no need for the ECB to make policy on these topics autonomously” (p. 25), which clearly means, in the context of their paper, that the ECB would merely act as a printing press. Monnet and van ’t Klooster (2023) give the example of the Deutsche Bundesbank giving exemptions to its discount policy in favour of export credit in the 1970s and 1980s as evidence that an independent central bank can implement a credit policy. However, the authors do not put these decisions in the context of the Bundesbank monetary policy framework. In fact, as in almost all other developed economies, the discount rate provided the floor to money market rates, implying that access to the discount window had to be rationed and that the discount rate moved in tandem with the Lombard rate, which was the ceiling to money market rates. The refinancing of export credit was thus not sheltered from interest rate hikes.

Monnet and van ’t Klooster (2023) also mention the Banque de France as an example of a central bank using credit policy in favour of exports, which can be interpreted in two ways. Firstly, there must have been economic pressure on the Bundesbank to do the same, especially in recession periods, since France and Germany are each other’s main economic partner. Secondly, since the Banque de France was made independent only in 1993, this tends to show that the Bundesbank’s independence should perhaps not be overrated. To be fair, Monnet and van ’t Klooster (2023) suggest that the creation of credit councils would certainly not fully suppress the independence of the central bank. However, it would curtail the prerogatives of the central bank, which would amount to a partial suppression of its independence. Finally, Kedward et al. (2022a, 2022b) do not even refer to the notion of central bank independence. This makes their approach more consistent than those of the other proponents of the measure, but also more disconnected from reality.

Issues

From a logical point of view

The proponents of the measure implicitly assume that the central bank would have better information on the climate risks incurred by banks than the banks themselves (of course they also implicitly assume that, for some reason, it should not share this information with the banks). However, as Dudi et al. (2021) put it: “In the absence of a consistent definition of environmental sustainability and of a reliable system of verification, it is unclear how to ensure that the fungible funds provided by banks are correctly and effectively used by individual borrowers to finance green projects” (p. 153). In other words, we cannot at the present stage identify what green or brown credit is, a point supporters of the proposed measure do not mention. Furthermore, there is no necessary negative relation between banks’ exposure to brown firms and investment to limit carbon emissions (Pfister and Valla, 2021). In fact, the relation could even be positive, with the more carbon-emitting firms needing to make the largest investments in order to reduce their carbon footprint and thus needing more support from banks.

If the ECB is to subsidise the fight against climate change, should it not also subsidise loans that supposedly create employment and increase productivity, thereby weighing on inflation when it is high, and conversely lower productivity when inflation is low, support diversity and reduce inequalities?⁵ This would transform the central bank into a development bank. To avoid such an outcome, secondary objectives of monetary policy would have to be ranked, which supporters of the proposed measure do not do. Monnet and van ’t Klooster (2023) do suggest such a ranking, in which the fight against climate change would come first, as they write: “the failure of central banks to consider the environmental impact of their instruments can undermine the broader role for monetary policy in supporting financial stability, government economic policy, stable employment and other central bank objectives” (p. 1). However, they do not justify this assertion, which thus seems to reflect their own personal judgement. In fact, excluding supposedly brown firms from access to green TLTROs (or for that matter corporate bond purchases) could create a stigma effect. This could lead these firms either to lay off workers or even to default, implying an increase in unemployment, and thus creating a conflict between the two secondary objectives of fighting climate change and supporting full employment.

From an economic point of view

As explained above, especially in the case of Kedward et al. (2002a, 2022b), the proponents of a green credit policy tend to downplay the role of market adjustments.

Furthermore, the proposed measure amounts to a fiscal transfer. Indeed, the same policy could be implemented by government agencies or development banks, which would distribute credit to finance the deserving projects with a subsidy financed by the budget allowing for the lowering of the cost of credit. The costs of this policy would be discussed by the legislature, which would be far more democratic than a discussion in a credit council (see above). The consequences for public finances would also be roughly identical, since the profit made by the ECB is distributed to the member states of the participating countries. It is telling that, in an article on loans distributed in the United States through government-backed direct loan and loan guarantee programmes – most notably from student loan programs Fannie Mae and Freddie Mac, and the Federal Housing Administration – Lucas (2016) refers to “credit policy as fiscal policy”. The proposed measure does not complement government action but rather replaces it. This could also raise a legal issue.

In the case of Monnet and van ’t Klooster (2023), as hinted above, the proposed measure is also part of a non-monetary approach to inflation. At least since the first oil shock, now more than half-a-century ago, it has been widely accepted that monetary policy’s role is to smooth aggregate demand and that there is little it can do against supply shocks, other than try to avoid a slippage in inflationary expectations, in the form of, e.g. a wage-price spiral. By contrast, within an approach of monetary policy that is reminiscent of the thesis of cost-push inflation in vogue until the early 1970s, the authors refer to credit in favour of green projects as disinflationary credit policy. However, this is a fallacy: according to the same logic, in the face of a positive wage shock, the authors might as well recommend that the ECB distribute credit to firms at below-market rates to avoid price increases.

The administrative costs of implementing the proposed measures could be very high. They would be incurred at the level of firms, banks and the central bank. These costs would result from the collection of appropriate information, of declaring and processing it, and of controlling both the allocation of credit to genuinely green projects (i.e. projects undertaken by firms which have a credible commitment to net-zero carbon emissions) and the pass-through to borrowers of privileged interest rates. To avoid the gaming of the measure, borrowing firms being granted the funds at privileged conditions would need to be checked in order to ensure that they do not have, nor will they have in the future, other projects that have a detrimental effect on the climate that they finance with other sources of funds. In fact, the administrative costs of implementing the measure could make the distribution of green credit more costly at the social level than that of refinancing banks using standard refinancing operations, thus eliminating the need for the reform and suggesting that other, more efficient ways to fight climate change should be implemented. Increasing the spread with the policy rate to compensate for high administrative costs would in turn increase the risk that banks and borrowing firms capture rents, which could in turn justify stepping up surveillance measures, thereby increasing administrative costs. By contrast, a carbon tax would dispense with the collection of data at the micro level, instead verifying the fulfilment of the commitment to net-zero carbon emissions at the aggregate level of countries, as is currently the case (Pfister & Valla, 2021).

Finally, one can question whether there really is a need for low interest rates to foster the fight against climate change. In the public debate, investment in decarbonisation is often characterised as a supplementary investment (i.e. it does not displace other investments). As a result, this investment is the marginal one, envisaged only when all other (by construction more profitable) investments have been undertaken. Rather paradoxically, this view is also held by the proponents of the measure. However, investment in the fight against climate change would be very beneficial from a social point of view. Consequently, the real issue is to align private interests with the social one. This is the starting point of the thrust of the economic literature on the subject and can be done in the most neutral and efficient manner through the imposition of a social price for carbon, as has been demonstrated a long time ago (Nordhaus, 1977). In line with this approach, banning securities issued by member states that do not implement policies consistent with the Paris Agreement on climate change from Eurosystem’s portfolios, as suggested by Pfister and Valla (2021), would arguably be a much more powerful measure than the proposed one, even though it could be more challenging to implement.

From a legal point of view

To start with, what supporting other EU policies means exactly should be clarified. Is it taking these policies into account when making decisions or serving them directly? Indeed, at the possible cost of oversimplifying, two different approaches to the contribution of all EU institutions to its objectives can be envisaged. The first one is “Napoleonic”, with all institutions implementing the policies determined by the centre. The second one relies on checks and balances, with the different institutions not opposing common objectives within their own realm, and contributing directly to them only when this does not infringe on the prerogatives of the other institutions. The proposed measure belongs clearly to the first approach.

It may also create difficulties with so-called monetary financing and conformity with market principles.

In relation to so-called monetary financing, Article 123(1) TFEU states:

Overdraft facilities or any other type of credit facility with the European Central Bank or with the central banks of the Member States (hereinafter referred to as “national central banks”) in favour of Union institutions, bodies, offices or agencies, central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of Member States shall be prohibited, as shall the purchase directly from them by the European Central Bank or national central banks of debt instruments.

This seems unrelated to the proposed measure, but the above has shown that green credit policy could very well and more democratically be implemented at the member state level. Thus, should the ECB take on responsibilities that can be assumed by member states, it could be seen as circumventing Article 123(1).

In relation to conformity with market principles, Article 127(1) TFEU states: “The ESCB shall act in accordance with the principles of an open market economy with free competition, favouring an efficient allocation of resources, and in compliance with the principles set out in Article 119”. However, Drudi et al. (2021) suggest that potential green TLTROs could raise level playing field issues for participating banks due to their differing ability to obtain and disclose relevant information as well as cross-country differences. In fact, these operations could also raise level playing field issues for firms that would be expected to benefit from the pass-through of subsidy interest rates granted to the banks having access to TLTOs.

Conclusion

There is a risk that the proposed measure might be well received by central banks, to the extent that the current context might be that of “mission creep” (Cochrane, 2020). The ECB announced on 30 January 2024 its intention to step up climate work with a focus on the green transition, climate and nature-related risks, and to explore the case for further changes to its monetary policy instruments and portfolios in view of this transition (ECB, 2024a). On 13 March 2024, the ECB announced changes to the operational framework for implementing monetary policy, including its intention to introduce new structural longer-term refinancing operations and a structural portfolio of securities (ECB, 2024b). On 27 March 2024, Elderson (2024) recalled that the ECB had in place a methodology to tilt the purchase of corporate bonds (but not that of public bonds) towards issuers with a better climate performance and that it was looking at setting limits on the share of assets issued by entities with a large carbon footprint. He also mentioned that climate change-related considerations would be incorporated into the design of future structural monetary policy operations, without specifying how this could be done.

If the latter measure were to amount to the introduction of green TLTOs, and thus incorporate the provision of liquidity at below-market rates, there would be a risk of a loss of credibility of the central bank. Indeed, the ECB would then interpret its contribution to EU policies in an extensive manner and intervene deliberately in the allocation of resources in the economy. This could be seen as a partial return to practices that were common in a period of high financial repression (Reinhart, 2012), when most central banks were not independent. In turn, this potential loss of credibility could contribute to making the ECB’s policies, particularly its monetary policy, less efficient and thus result in a loss of social welfare. In other words, adopting the proposed measure would be going one step too far.

• 1 China, for instance, is not necessarily a model to follow, particularly as far as the fight against climate change is concerned. Furthermore, the People’s Bank of China submits all sorts of credit, and not just green credit, to interest rate ceilings and floors.
• 2 The European System of Central Banks (ESCB) comprises the European Central Bank (ECB) and the National Central Banks (NCBs) of the member states participating in the euro area.
• 3 In the case of differentiated deposit rates, the rates would have to be higher for green banks in order to reward them. However, by increasing the level of market interest rates in the current context of excess liquidity supply, this would be tantamount to a tightening in the monetary policy stance, a point which seemingly escapes the attention of the authors.
• 4 This proposal is largely inspired by the example of the French Conseil National du Crédit (CNC). The CNC was a government agency, created in 1945, as a part of an act that also nationalised the Banque de France and the main French banks. It was chaired by the Ministry of Finance and composed of nearly fifty members, most of them being civil servants and representatives of banks and their employees, with five members from the legislative (National Assembly and Senate) and consultative (Economic and Social Council) assemblies. It had a chiefly consultative role, but also had, until 1984, a limited jurisdictional and regulatory power. It was abolished in 2003.
• 5 De facto, Monnet and van ’t Klooster (2023) suggest thet “it is desirable to protect some sectors during a tightening cycle because certain types of investment prevent, rather than cause, inflation” (p. 1). In the same vein, Colesanti Senni et al. (2023) suggest that “the ECB could also link its refinancing operations to more targeted programs, such as the May 2022 REPowerEU action plan, which seeks to address the climate-crisis while also ending the EU’s dependence on Russian fossil fuels” (p. 25) as well as “targeted housing and energy programs” (pp. 26-27).