VoxEU Column Environment Financial Markets

Brown assets might be the next subprime

The momentum towards greening the economy implies transition risks that represent new threats to financial stability. The risk of a run on brown assets, similar to that seen during the subprime crisis, can have widespread destabilising effects. This column proposes a liquidity backstop with an access fee proportional to carbon emissions, and a borrowing rate independent of emissions. It argues that such a facility will help green the economy, re-establish production efficiency, and avoid highly inefficient runs. An orderly reallocation of capital to a greener economy can lift long-term growth and facilitate the post-Covid recovery.

Finance is waking up to global warming. While governments have failed to coordinate on taxing carbon emissions, investors are increasingly expressing preferences for portfolios that meet some minimum environmental, social, and governance (ESG) standards.1 Such market forces can become a powerful driver to a greener economy. 

As asset managers are increasingly seeking means to signal the rebalancing of their financial positions toward firms with low carbon emissions, ‘best in class allocations’ applied to carbon footprint imply dumping or staying away from financing the brownest firms. An important argument speaking in favour of this rebalancing strategy is that carbon emissions are extremely skewed across firms. The top 1% of most polluting firms account for 40% of total Scopes 1 to 3 carbon emissions in 2018 (Figure 1, see Ehlers et al. 2020). Therefore, asset managers can effectively reduce the carbon footprint of their portfolio by excluding (or reducing their exposure to) a small enough number of firms such that the impact on the financial performance and tracking error of their portfolio remains limited (Andersson et al. 2016, Bolton and Kacperczyk 2021, Jondeau et al. 2021b). 

The risk that brown assets generate a crisis similar in its working to the subprime mortgage crisis is material. Indeed, the two components of the subprime crisis are already in the making. Talking down brown assets will likely lead to a ‘carbon stigma’, very much like collateralised debt obligations (CDOs) suffered at some point from a ‘subprime stigma’. The second component is the uncertainty surrounding the degree of acceptability (either by society, investors, or regulators) of the level of carbon emissions in portfolios of corporates.

The parallel between brown assets and the market for subprime mortgages is striking. The brownest assets (Figure 1) are a small proportion of the entire capital in the economy, just like the worst subprimes were. The latest vintages of subprime-based securities had the worst credit risks with some borrowers not paying even the first mortgage instalment. Early vintages had much lower levels of credit risk. Likewise, many corporates among utilities and electricity production, while not among the worst polluters, are still emitting high levels of carbon. They may very well be also subject to the same carbon stigma as the worst emitters in their sector. 

When it comes to stigma on an asset class, financial markets can lose their ability to screen risks. Once exposure to subprime became suspicious, even the early vintages of subprime-based collateralised debt obligations became dubious, and investors ran on the entire market; they stopped lending to institutions that were thought to be too exposed to the mortgage market. As the subprime crisis made it all too clear, compounding a stigma with uncertainty on the degree of exposure adds up to a time bomb of financial instability.

Figure 1 Absolute carbon emissions by carbon intensity percentile, FY2018


A run could discipline firms, but it is still inefficient 

To be sure, the ‘run’ phenomenon could be seen as the Damocles sword hanging over brown firms and asset managers. It could entice decision makers to switch to greener production processes. Yet the increase in transition risk due to the stigma on stranded assets should be managed.

A run is a very coarse market mechanism to discipline firms. As such, and on top of the collateral damage it can cause, it does not consider the two facets of carbon emissions: a firm can produce a lot of carbon because it produces on a large scale, or because its production process is inherently carbon-intensive. While they both look brown, these firms should be treated differently when transitioning towards meeting environmental standards. 

In Jondeau et al. (2021a), we analyse a production economy where firms produce with different intensity of carbon emission. For example, for every watt of electricity it produces, a water-based electricity company has a lower carbon intensity than a coal-based electricity company. In addition, firms can all undertake some costly investment to reduce their effective emissions from their initial levels. Under fairly general conditions, efficiency requires that no firm is entirely excluded from the market. Firms should operate at different scales, with the worst polluters cutting their production more. And all firms should invest to reduce emissions from their initial levels. 

We consider the existence and implications of runs on brown assets. When investors run on what they believe to be brown firms, the production of these firms has to stop, leaving some assets stranded. This is inefficient given the firms may have already made some initial investments. Also, investors may coordinate on a socially inefficient equilibrium where nothing is green enough, running even on firms that are relatively green and that should optimally keep operating. This is also inefficient. Runs are also inefficient because investors who focus strictly on the run risk cannot price the carbon emission externality. Firms that end up greener than the ad hoc threshold chosen by investors can carry on polluting, while taking into account the social preference for lower carbon emission should entice them to reduce carbon emissions. 

Figure 2 shows, for an economy where people dislike pollution, the socially optimal investment in each firm with increasing levels of emission intensity, θ, typically the amount of carbon per dollar of output of the firm, as well as the level of emission reduction for each firm. Importantly, the planner invests in each firm so that the marginal product of capital net of the cost to reduce emissions is the same across all firms. 

Figure 2 Socially optimal investments


We show that a well-designed liquidity backstop facility can address the inefficiencies we outlined above. It will help firms that are subject to a run and prevent assets from becoming stranded. Notably, we find that the lending rate charged to firms borrowing from the liquidity backstop facility should not depend on the level of carbon emissions of those firms. However, the regulator should charge an access fee that depends on carbon emissions.2 

Collateral frameworks and the link to other central bank policies

The reasoning applies either to non-financial corporates or to the portfolio of financial firms. The carbon footprint of the portfolio will depend on its proportion of brown firms. The stigma and risk of runs may very well apply to these financial intermediaries. Given the vast cross-exposure of financial intermediaries, notably on money markets, brown stigma or the fear of brown stigma can induce in turn a systemic market freeze similar to the one the relatively small subprime market led to in 2008. 

In this respect, our prescribed policy is a theoretical foundation to design collateral frameworks that take into account the risks that brown stigma undermines financial stability. 

It is worth mentioning that the ECB has recently decided that bonds with coupon structures linked to the satisfaction of some sustainability performance targets will be eligible as collateral for Eurosystem credit operations and also for Eurosystem outright purchases for monetary policy purposes (provided they comply with all other eligibility criteria). This evolution of the ECB collateral framework can facilitate an orderly reallocation of capital to a greener economy. As stressed by De Haas et al. (2021) finance makes a difference on the transition of corporates to greener technology, which, as argued by Benassy-Quéré et al. (2020), can both lift the long-term growth of Europe and give a perspective to its post-Covid recovery. 


Andersson, M, P Bolton and F Samama (2016), “Hedging climate risk”, Financial Analysts Journal 72(3): 13-32.

Bénassy-Quéré, A, R Marimon, P Martin, J Pisani-Ferry, L Reichlin, D Schoenmaker and B Weder di Mauro (2020), “Repair and reconstruct: A Recovery Initiative”,, 20 April.

BlackRock (2020), “Blackrock survey shows acceleration of sustainable investing”, Press Release, 3 December. 

Bolton, P and M Kacperczyk (2021), “Global pricing of carbon-transition risk”,, 24 March.

De Haas, R, R Martin, M Muûls and H Schweiger (2021), “Barriers to net-zero: How firms can make or break the green transition”,, 19 March.

Ehlers, T, B Mojon, F Packer and L A Pereira da Silva (2020), “Green bonds and carbon emissions: Exploring the case for a rating system at the firm level”,, 12 December.

Jondeau, E, B Mojon and C Monnet (2021a), “Greening (runnable) brown assets with a liquidity backstop”, Bank for International Settlements Working Paper No. 929.

Jondeau, E, B Mojon and L A Pereira Da Silva (2021b), “Excluding firms with large carbon emission for portfolio investment”, Bank for International Settlements Working Paper.

Mercure, J F, H Pollitt, J E Viñuales, N R Edwards, P B Holden, U Chewpreecha, P Salas, I Sognnaes, A Lam and F Knobloch (2018), “Macroeconomic impact of stranded fossil fuel assets”, Nature Climate Change 8(7): 588-593. 

Oustry, A, B Erkan, R Svartzman and P F Weber (2020), “Climate-related risks and central banks’ collateral policy: A methodological experiment”, Banque de France Working Paper No. 790.


1 BlackRock (2020) reports that 88% of its investors are expressing a preference for environmental, social, and governance (ESG) portfolios with the environmental pillar being a much bigger concern than either social or governance pillars.

2 Similarly, deposits insurance, as implemented by the Federal Deposit Insurance Corporation (FDIC) in the US, relies on a mechanism in which pricing bears resemblance to the pricing of the liquidity backstop facility we propose: banks have to satisfy some liquidity and reserve requirements to benefit from the insurance mechanism. Once the bank is insured, the FDIC deposit insurance covers all deposit accounts (up to a certain limit) irrespective of the deposit base of banks and bank risk.

2,310 Reads