Network for Greening the Financial System
Technical document
Adapting central bank
operations to a hotter world
Reviewing some options
March 2021
NGFS REPORT
2
T
he pandemic and its fallout have fast-forwarded us into a new dimension of central bank support to our economies. Central banks across
the globe have shown unprecedented levels of resolve, responding swiftly and exibly using a wide array of monetary instruments.
At the same time, climate change remains an urgent and fundamental threat to our prosperity and collective well-being. Unlike for the pandemic,
however, in the climate crisis we cannot see light at the end of the tunnel. The urgency to act is greater than ever: climate risks no longer lie beyond
the horizon, they are already materializing. The time to take action is now.
The NGFS started 2021 with undiminished energy and vigor. Our members are more determined than ever to get active, pressing ahead with
concrete proposals on how to better account for climate-related risks in central banking and banking supervision. We strongly believe that
now is the time for central banks to seriously consider how the progress made in reecting climate-related risk in supervisory and macro-
prudential methods can be matched by similar steps in monetary policy operations.
The report “Adapting central bank operations to a hotter world” examines the implications of climate change for central banks’ operational
frameworks and for the implementation of monetary policy in practical terms. Building on a common understanding among NGFS members
that climate change has implications for the conduct of monetary policy, this report oers the most comprehensive analysis to date. Practitioners
from the central bank community reviewed collateral and counterparty policies, asset purchases and credit operations with a view to oering
a menu of options for climate-related adjustments in more concrete terms.
This report does not prescribe a particular course of action. Regardless of their specic roles and mandates, central banks ought to be aware of
climate risks that could threaten the integrity of their balance sheets. However, each central bank needs to decide for itself the best way to reect
climate risks in its operational framework. We are sure that this report will oer invaluable guidance for central banks in making these strategic
choices with regard to their monetary policy operations.
Of course, this is only the beginning. More work is needed to overcome obstacles and to fully integrate climate-related considerations into
monetary policy. These issues will rank high among NGFS priorities going forward.
Central banks clearly need to play their part in the joint global eorts to curb climate change as an urgent and universal challenge. While we
cannot take on the tasks of governments, we also cannot be mere bystanders in the transition to a net zero economy. It is our responsibility
to take on the challenge we are facing as publicly accountable institutions, serving our societies.
We are grateful to all NGFS members and observers for contributing to our common cause in a truly challenging environment. Our network is
thriving thanks to your determination and ideas and we urge you to stay committed. Our special thanks go to the lead authors of this report and
its contributors, as well as the NGFS secretariat. Their tireless eorts have made it possible for us to mark this important milestone for the NGFS.
Joint foreword by Frank Elderson and Dr Sabine Mauderer
Dr Sabine Mauderer
Chair of the workstream “Scaling up Green Finance”
Frank Elderson
Chair of the NGFS
NGFS REPORT
3
Table of Contents
Executive summary 4
1. Introduction 9
2. The state of play 12
2.1. Climate change brings new financial risks for central banks 12
2.2. Adapting traditional central bank models to climate change? 13
2.3. Potential courses of action 14
3. Principles for assessing potential climate-related adjustments
to monetary policy operational frameworks 16
3.1. Consequences for monetary policy effectiveness 16
3.2. Contributions to mitigating climate change 16
3.3. Effectiveness as risk protection measures 16
3.4. Operational feasibility 16
4. Reviewing potential options 18
4.1. Identifying the options 18
4.2. Summary assessment 19
4.3. Open questions 21
5. Disclosure 22
5.1. Is disclosure a prerequisite for other potential adjustments? 22
5.2. Requiring disclosure from eligible collateral issuers and/or monetary policy
counterparties 25
5.3. Disclosing the central bank’s own exposures to climate-related risks 25
6. Strategic choices when dealing with climate change 27
6.1. Risk tolerance and assessment 27
6.2. Metrics 28
6.3. Data 29
6.4. Balancing trade-offs 30
7. Annexes 31
Annex 1. Detailed review of options 31
Annex 2. Climate-related metrics 46
Annex 3. Coordinating climate-related adjustments to operational frameworks 48
Annex 4. Bibliography and overview of recent proposals 50
8. Acknowledgements 54
NGFS REPORT
4
The context calls for concrete action
Under all possible scenarios, climate-related risks will have
consequences for the economic outlook, for the financial
system in which central banks operate and, thus, for the
conduct of monetary policy. The timing and severity of
these consequences depend on how swift and effective
transition policies are.
Moreover, climate change poses new financial risks to
central banks’ monetary policy operations. Climate-related
financial risks could impact directly on both central bank
counterparties and the financial assets used in monetary
policy operations (as collateral for credit operations or for
outright purchases).
As a result, climate-related shocks could generate financial
losses for central bank balance sheets and, in extreme cases,
they could affect the smooth implementation of monetary
policy by exposing various monetary policy transmission
channels to the impacts of physical and transition risks.
Central banks can adapt their monetary
policy operational frameworks to reect
climate-related risks
Governments have a much broader and more effective
range of tools and policies available to prevent and mitigate
climate-related risks than central banks, and they are the
actors responsible for designing and conducting national
and international climate policies.
However, contingent on their mandate, central banks have
a responsibility to review their operational frameworks to
ensure they remain resilient to emerging climate-related
risks and to safeguard the continued smooth conduct
of monetary policy, i.e. to consider the effect of climate-
related risks on their operations as well as the effects of
their actions on exposures of other entities, including the
financial sector, to climate-related risks.
There is a broad consensus among members of the Network
for Greening the Financial System (NGFS) that, at the very
least, central banks should carefully assess, and where
appropriate adopt, additional risk management measures
to protect their balance sheets against the financial
risks brought about by climate change. However – and
reflecting the diversity of existing central bank operational
frameworks – there is as of yet no consensus among central
banks as to what climate-related adjustments would be
optimal. Identifying the relevant measures and assessing
the adequate level of protection against climate-related
financial risks, and the quantification thereof, is a challenge
for central banks at the current juncture.
Where it falls within their policy remit, central banks
could also consider going beyond the adjustment of their
operational frameworks solely from a risk management
perspective by seeking to ensure that their monetary
policy operations do not undermine the transition to a
low-carbon economy and/or by exploring ways in which
they can actively support that transition.
In practice, the frontier between these alternative
approaches (mitigating balance sheet risk on the one hand,
and actively supporting transition on the other) is blurred
and may depend on the actual calibration of operational
measures as well as the central bank’s mandate.
According to current scientific evidence, taking no action
is not viewed as a sustainable option given the systemic
impacts of climate change on the real economy, on financial
risk, on market prices and thus on the conduct of monetary
policy and on monetary policy frameworks. At the same
time, central banks need to be mindful about the potential
risk involved in considering adjustments based on what
is still a limited body of information, which may have an
impact on their credibility.
The menu of options available to central
banks to factor climate-related risks into their
operational framework is potentially large
Adjustments could be considered across the main operational
functions that central banks carry out for the purposes of
implementing monetary policy. This report analyses possible
changes to three of the most important policy fields: credit
operations, collateral policies, and asset purchases.
Executive summary
NGFS REPORT
5
The review concentrates on potential measures on the asset
side of a central bank’s balance sheet. Hence, the stylised
options listed in Table 1 all pertain to liquidity-providing
instruments.
Based on the available literature and expert analyses, the
review by the NGFS group of experts focuses on nine stylised
options across these three main policy fields (Table 1).
They were chosen because they are relevant to multiple
central banks and relate to existing tools. Some options
represent a greater departure from standard central bank
operational policies than others.
Depending on their mandate, legal environment and
individual assessment, certain central banks may
not find some of the stylised options to be feasible.
The review therefore contains neither recommendations,
nor indications of members’ preferences.
Table 1. Selected stylised options for adjusting operational frameworks to climate-related risks
Credit operations
a
(1) Adjust pricing to reflect
counterparties’ climate-related
lending
Make the interest rate for central bank lending facilities conditional on the extent to which a
counterparty’s lending (relative to a relevant benchmark) is contributing to climate change mitigation
and/or the extent to which they are decarbonising their business model.
(2) Adjust pricing to reflect the
composition of pledged
collateral
Charge a lower (or higher) interest rate to counterparties that pledge a higher proportion of low-carbon
(or carbon-intensive) assets as collateral or set up a credit facility (potentially at concessional rates)
accessible only against low-carbon assets.
(3) Adjust counterparties’ eligibility Make access to (some) lending facilities conditional on a counterparty’s disclosure of climate-related
information or on its carbon-intensive/low-carbon/green investments.
Collateral
b
(4) Adjust haircuts
c
Adjust haircuts to better account for climate-related risks. Haircuts could also be calibrated such that
they go beyond what might be required from a purely risk mitigation perspective in order to incentivise
the market for sustainable assets.
(5) Negative screening Exclude otherwise eligible collateral assets, based on their issuer-level climate-related risk profile for
debt securities or on the analysis of the carbon performance of underlying assets for pledged pools
of loans or securitised products. This could be done in different ways, including adjusting eligibility
requirements, tightening risk tolerance, introducing tighter or specific mobilisation rules, etc.
(6) Positive screening Accept sustainable collateral so as to incentivise banks to lend or capital markets to fund projects
and assets that support environmentally-friendly activities (e.g. green bonds or sustainability linked
assets). This could be done in different ways, including adjusting eligibility requirements, increasing risk
tolerance on a limited scale, relaxing some mobilisation rules, etc.
(7) Align collateral pools with
a climate-related objective
Require counterparties to pledge collateral such that it complies with a climate-related metric at an
aggregate pool level.
Asset purchases
d
(8) Tilt purchases Skew asset purchases according to climate-related risks and/or criteria applied at the issuer or asset level.
(9) Negative screening Exclude some assets or issuers from purchases if they fail to meet climate-related criteria.
a Credit operations are widely used to provide aggregate liquidity and usually take the form of collateralised lending.
b Collateral policy denes the range of assets that can be pledged to secure central bank credit operations, as well as the risk control measures that apply
to them.
c Annex 1 expands upon the dierent approaches for haircuts and valuation adjustments.
d Central banks may buy a variety of assets from both public and private sectors, typically in an eort to exert greater inuence on longer-term interest rate
levels and spreads while improving market liquidity.
NGFS REPORT
6
Four criteria can help review the menu
of options available to central banks
Assessing different climate-related adjustments to monetary
policy operations is difficult because of the heterogeneity
of central bank operational frameworks.
Regardless of these differences, the potential adjustments
to central bank operations can be assessed against four
general principles (see Table 2). These are: (1) Consequences
for monetary policy effectiveness; (2) Contributions
to mitigating climate change; (3) Effectiveness as risk
protection measures; and (4) Operational feasibility.
Depending on their mandate and on the course of action
chosen, central banks may assign different weights to
these four principles.
Consequences for monetary policy effectiveness.
Assessing the implications for the effectiveness of
monetary policy operations (including in terms of lending
or purchasing capacity by the central bank, potential
distortions, stigma, etc.) of any of the options is challenging
since they very much depend on their exact design as well
as the central bank’s specific circumstances. Still, options
which materially reduce available monetary policy space,
or which can jeopardise the efficacy of monetary policy,
are unlikely to be considered desirable, in particular if
their design and calibration cannot be used to minimise
any unintended consequences. While further jurisdiction-
specific work is needed, a few preliminary points can be
made. Some options run the risk of curtailing, more or less
significantly, central bank operations and the policy space.
These options include (i) negative screening that would
(a) exclude a significant number of counterparties from
credit operations based on their carbon footprint or carbon
disclosure; (b) exclude assets potentially representing a
significant share of the purchasable universe or of eligible
collateral; and (ii) adjusting the pricing of credit operations
to the composition of collateral. For other options, the
implications for the effectiveness of monetary policy may be
less relevant or even negligible, though this ultimate impact
would need to be assessed in light of each central bank’s
circumstances. Another key point of vigilance concerns the
potential unintended consequences that some options
may have for financial stability.
Contributions to mitigating climate change. A few
options may be more impactful from a climate change
mitigation perspective than others. These include
measures aimed at (i) adjusting the pricing of targeted
credit operations to a lending benchmark; (ii) positively
screening collateral; (iii) aligning collateral pools; and
(iv) tilting asset purchases. They typically consist of
modifying existing tools without fully overhauling their
design (e.g. leveraging pricing schemes for targeted credit
operations) in order to encourage lenders to originate
or invest more in low-carbon and transition assets. It is
unlikely that they would materially curtail operations and
policy space. Seen from this perspective, they would be
consistent with the smooth implementation of monetary
policy but still technically challenging to operationalise.
At the same time, potential implications for asset pricing
and market functioning have to be carefully assessed.
Eectiveness as risk protection measures. Many of the
options reviewed would probably better shield central
bank balance sheets against increasing financial risks, most
effectively through those options aimed at directly reducing
risk exposure (to issuers or counterparties). Accordingly,
the following options are viewed as being probably
risk-protective: (i) negatively screening counterparties
to credit operations based on their carbon footprint or
carbon disclosure; (ii) adjusting haircuts and valuations;
(iii) negatively screening collateral; (iv) aligning collateral
pools; (v) tilting asset purchases; and (vi) negatively
screening purchasable assets. However, for some of them
(e.g. negative screening options), this potentially positive
impact could be diluted, or in some cases outweighed,
if the reduction in the eligible universe were associated
with higher financial risk concentrations, or greater credit
risk unrelated to climate change. A priori and contingent
on each central bank’s mandate, options designed from
a financial risk perspective may be less exposed to legal
risks and challenges than others designed to support
climate-related objectives, especially if the latter are
seen as subsidising some economic sectors, issuers or
assets. Yet for many options, the actual impact from a
risk protection perspective is difficult to assess without a
detailed specification.
Operational feasibility. All options entail significant
changes to central bank operational frameworks. The
least challenging options to operationalise are the
least sophisticated ones (e.g. the simplest form of
exclusion measures) in terms of addressing climate-
related risks. Conversely, the options that are less likely
NGFS REPORT
7
to entail adverse consequences for monetary policy
effectiveness are typically associated with somewhat
higher operational complexity. This is the case for
(i) adjusting the pricing of targeted credit operations
to a lending benchmark; (ii) adjusting haircuts and
valuations; (iii) aligning collateral pools; and (iv) tilting
asset purchases. Whether any additional complexity
would be warranted to achieve a reduction in financial
risk or improved climate outcomes would need to be
assessed on a case-by-case basis.
Table 2. Simplied comparative assessment of the selected generic options under review
ADJUSTING
PRICING TO
LENDING
BENCHMARK
ADJUSTING
PRICING TO
COLLATERAL
ADJUSTING
COUNTERPARTIES’
ELIGIBILITY
HAIRCUT
ADJUSTMENT
NEGATIVE
SCREENING
POSITIVE
SCREENING
ALIGNING
COLLATERAL
POOLS
TILTING NEGATIVE
SCREENING
CONSEQUENCES FOR
MONETARY POLICY
EFFECTIVENESS
CREDIT OPERATIONS
COLLATERAL
ASSET PURCHASES
CONTRIBUTION TO
MITIGATING CLIMATE
CHANGE
EFFECTIVENESS AS RISK
PROTECTION MEASURE
OPERATIONAL
FEASIBILITY
POTENTIAL IMPACT :
STRONGLY POSITIVE MINIMAL STRONGLY NEGATIVE
POSITIVE NEGATIVE
(1) (2) (3) (4) (5) (6) (7) (8) (9)
The assessment is based on qualitative expert judgement, and more formal quantitative analysis may be needed. It aims to guide the reader through
the report and should not be interpreted as recommending any measure. Colour-coding is used to avoid any “netting” across criteria. The table uses
a limited number of colours for reasons of simplicity. More nuanced analyses of options are provided in Annex 1.
All in all, adjusting central bank operational frameworks
to more adequately reflect climate-related considerations
is feasible. Yet the climate-related adjustments of central
bank operations have to overcome a range of practical and
analytical challenges, including data gaps and uncertainties
with regard to risk quantification. There is a priori no “one
size fits all” option that clearly maximises the benefits across
all four principles listed above.
Enhanced disclosure of climate-relevant
data is instrumental to support
central banks’ actions
Enhancing the disclosure of climate-relevant data is a policy
issue that cuts across many of the potential options, while
disclosure requirements may be designed by central banks
depending on their respective responsibility within their
jurisdiction. Increasing the quantity and quality of climate-
relevant information is a critical step in enabling central
banks and market participants to better understand their
exposures to climate-related risks.
In some cases, the increased availability of climate-related
information may be a prerequisite for adjusting certain
operational frameworks, especially where operational
changes may pose legal and reputational risks. However,
some climate-related adjustments to operational
frameworks can be developed in parallel to initiatives
fostering comprehensive data disclosure. When balancing
the need for robust and comprehensive data against
the opportunity cost of inaction, central banks should
be cognisant of the risk that acting early with imperfect
information could be less costly than acting only once
stronger data standards have emerged.
Introducing disclosure requirements in monetary policy
operations could help foster harmonised, transparent,
reliable and comparable data. To reduce the operational
burden of disclosure requirements and cater for issues
associated with comparability and transparency, central
banks could make use of existing reporting frameworks and
minimise deviations from such frameworks or forthcoming
regulations in their respective jurisdictions.
NGFS REPORT
8
Central banks may wish to disclose climate-related
information on their own policy operations and financial
activities. This could be motivated by considerations about
transparency and accountability to the public about the
climate-related risks they take as part of their operations.
It can also serve to signal a central bank’s commitment to
enhancing the availability of climate-related risk information
and set a positive example to assist market participants in
developing their own disclosure frameworks.
To take action, central banks must decide
on some strategic issues
Central banks can formulate a clear strategic view on their
tolerance of climate-related risks and decide how forward-
looking they wish their frameworks to be.
Central banks need to form a clear opinion surrounding
the appropriateness of various climate-related metrics in
order to adjust their operational frameworks. At the current
juncture, in the absence of reliable and commonly agreed
ways of putting a price tag on climate-related risks, central
banks wishing to act may have no choice but to consider
using non-financial climate-related metrics as a pragmatic
starting point.
Central banks should develop policies to monitor and
manage issues surrounding data quality and availability.
The limited availability and accuracy of relevant data is
currently constraining virtually all climate-related risk
metrics.
Figure 1. Strategic choices for adapting monetary policy operational frameworks to climate-related risks
STRATEGY
RISK MANAGEMENT
VS OTHER
FUNCTIONS
METRICS & DATA
MEASURES
HOW ARE CLIMATE CHANGE RISKS DEFINED?
WHAT IS CENTRAL BANKS’ TOLERANCE TOWARDS THEM?
TRADE-OFF: ACCURATE RISK ASSESSMENT VS SWIFT IMPLEMENTATION
EFFECTIVE MONETARY POLICY IMPLEMENTATION?
BACKWARD OR FORWARD-LOOKING METRICS?
NEED FOR DATA GATHERING
HOW ARE RISK MITIGATION MEASURES DESIGNED ?
Against this backdrop, central banks face some trade-offs
when dealing with climate-related risks. On the one hand,
central banks have to operate within their specific legal
framework, and as publicly accountable institutions,
they have to provide rigorous evidence in support of all
actions they take – this may lead them to taking a cautious
approach to adopting policies for new risk drivers such as
climate change. On the other hand, central bank balance
sheets might already be exposed to climate-related risks,
which is why early action to mitigate them would be
called for in the interests of the prudent risk management
of public funds. Owing to the heightened uncertainty
surrounding the exact timing and magnitude of climate-
related risks’ materialisation, the optimal policy for many
central banks is likely to be to adopt gradual, predictable,
precautionary risk protection measures. This approach
should be in line with, and conducive to, emerging best
practices.
NGFS REPORT
9
This report forms part of the work of the Network
for Greening the Financial System’s (NGFS) group
of experts that investigates the possible effects of
climate change on the conduct of monetary policy.
The first report, “Climate change and monetary policy –
initial takeaways” (NGFS, 2020a), explored how climate
change affects key macroeconomic variables and, as a
consequence, the conduct of monetary policy and its
transmission channels. Central banks were recommended
to consider the possible effects of climate change on the
economy and thus on the conduct of monetary policy.
To do so, they may need to reinforce their analytical,
forecasting and modelling toolkit so as to better capture
and understand the economic and financial impacts of
climate change. Moreover, they may evaluate whether
and how they might need to adapt their monetary policy
operational framework to climate change.
This second report focuses on the operational
implications of climate change for central banks, with
a particular focus on the implementation of monetary
policy. It is motivated by several considerations, which
are related to one another and on which further work is
needed.
First, in all possible scenarios, climate change will
impact on economic agents and their behaviour.
An orderly transition towards a 1.5°C-2°C of average global
temperature rise requires substantial mitigation measures
to reduce physical risk, which will require public, economic
and financial agents to invest and adapt. By contrast, a lack
of mitigation and adaptation policies would lead to a “hot
house world” scenario which is expected to result in rapidly
soaring costs stemming from spiralling physical risk impacts
(see Figure 2). Alternatively, there could be “disorderly”
transition scenarios – perhaps related to the effectiveness,
timing, heterogeneity and acceptance of mitigation
policies – in which a range of physical risks (limited or
high) could unfold. Under all scenarios, there could be swift
shifts in sentiment amongst financial market participants,
affecting asset valuations and increasing volatility in risk
perception. Financial markets could eventually witness a
flight into assets deemed safest from the standpoint of
climate change, and out of assets considered least safe
from that vantage point. The bottom line is that, in all
scenarios, the economic and financial ecosystem in which
central banks conduct their monetary policy will very likely
change, which has implications for the design of monetary
policy operational frameworks.
Figure 2. NGFS climate scenarios framework
Hot house world
We continue to
increase emissions,
doing very little, if
anything, to avert
the physical risks
Orderly
We start reducing
emissions now in a
measured way to
meet climate goals
Too little, too late
We don’t do enough
to meet climate goals,
the presence of
physical risks spurs a
disorderly transition
Transition pathway
Disorderly
Disorderly
Sudden and
unanticipated
response is disruptive
but sucient enough
to meet climate goals
Met Not met
Based on whether climate targets are met
Strength of response
Transition risks
Physical risks
Orderly
Source: NGFS (2019a)
Second, monetary policy transmission channels
1
are
likely to become increasingly exposed to climate-
related risks – that is, both physical and transition
risks. The credit channel could experience the greatest
effects, which may be a source of concern in countries
where it is the predominant transmission channel. More
generally, as the NGFS has already pointed out,
2
climate
change has the potential to affect financial intermediaries’
balance sheet capacity,
3
which could weigh on their ability
1 These comprise the interest rate channel, the expectations channel, the credit channel (via bank lending and market-based finance) and the risk-taking
channel.
2 See NGFS (2019b).
3 These include credit institutions, insurance companies, broker-dealers and different types of investment funds (pension, money market, mutual
funds). In this introduction we simply refer to “banks”.
1. Introduction
NGFS REPORT
10
to transmit monetary policy effectively to the broader
economy. Climate change can also affect monetary
policy transmission through the expectations channel.
Though climate-related risks might materialize later,
economic agents may anticipate them and adapt their
behaviour accordingly. This, in turn, could affect monetary
policy and its transmission channels. On the other hand,
the extent to which these potential effects could affect
the ability of monetary policymakers to achieve their
objectives is not yet known, and there is a consensus
that further work by central banks is needed on this front.
Therefore, central banks are each expected to carefully
assess whether those risks have material implications for
the implementation of monetary policy.
Third, because it will aect the net worth of economic
agents, climate change could reduce the value of the
assets available to banks to participate in central bank
monetary policy operations. The balance sheets of firms
and households may be hit – directly and indirectly – by
physical and transition risks. Both climate change and
new transition policies may affect the net present value
and probability of default of assets pledged to central
banks, and thus impact collateral values. The quantitative
importance of such effects still needs to be assessed. Lastly,
more frequent and more damaging extreme weather
events and changes to the regulatory environment for
greenhouse gas (GHG)-emitting sectors may affect asset
prices in the financial sector and the real economy alike.
The extent to which central banks may find it
appropriate or advantageous to adjust their existing
operational frameworks still needs to be assessed.
From a broad perspective, in recent instances where
central banks have intervened to reinforce an impaired
transmission of monetary policy (e.g. during the Great
Financial Crisis, 2007-09), their actions aimed to address
concrete and manifest financial market malfunctioning.
Climate change, by contrast, while already manifest,
represents a risk that will likely crystallise in such a way that
could disrupt the monetary policy transmission channel
in the future. Central banks need to assess, measure and,
where appropriate, manage the risks from climate change
just as they would for any other type of financial risk,
while safeguarding the continued effective transmission
and smooth implementation of monetary policy today.
In this endeavour, they need to assess whether, and take
into account that, a failure to make orderly and timely
adjustments to their monetary policy framework may
endanger their ability to meet their primary objectives of
monetary and financial stability in the future. Nevertheless,
while central bank policies can potentially complement
actions by governments to facilitate, manage and bring
forward climate transition, they cannot be a substitute
for climate policies.
To shed light on these issues, this report builds on
three inputs. First, it draws on a survey of NGFS member
central banks
4
that aimed to identify whether central banks
across the world are currently thinking of adjusting their
operational frameworks, and how, in order to take account
of climate-related risks. While this survey confirmed that
there is a growing shared awareness of the magnitude of
the climate challenge and the importance for central banks
of managing climate-related risks contingent on their
mandate, it also revealed that concrete action by central
banks has been limited. This likely reflects the systemic
nature of the challenges that climate change poses and
the complexity and novelty of measuring and modelling
those longer-term risks dynamically. Second, the report
leverages on an extensive review of studies and proposals
by researchers, academics and other non-central bankers
about the operational implications of climate change for
monetary policy (see Annex 4). That review shows the wide
range of monetary policy tools currently used by central
banks across the world, suggesting that adjustments to
those policy tools to address climate-related risks will need
to be tailored to each institution’s own circumstances.
Third, the report analyses case studies of climate-related
measures implemented by central banks (see boxes in
Annex 1). These illustrate the variety of options available
to central banks and objectives pursued.
The report should be read as a first attempt by central
banks to look jointly into the potential operational
implications of climate change for monetary policy
implementation. It does not contain any specific
recommendations. Rather, it seeks to identify the strategic
choices, general concepts and potential adjustments to
operational frameworks that central banks may wish to
4 See NGFS (2020b).
NGFS REPORT
11
consider, as well as the possible constraints on change
which need to be taken into account. Further economic
research and work by the central banking community
is needed for robust conclusions to emerge on several
points raised in this report. Besides, each central bank
is uniquely placed to assess whether and how climate-
related risks may affect the design of its own monetary
policy tools.
The focus of the report is on climate-related nancial
risks. These are referred to interchangeably as either
climate-related financial risks or as climate-related risks.
This report is organised as follows. Chapter 2 takes stock of
monetary policy operational frameworks to identify the key
constraints central banks face when considering adapting
them to climate-related risks. Chapter 3 presents four general
principles that could be used to analyse and compare
potential options for these adjustments. Chapter 4 reviews
a selected set of potential adjustments to operational
frameworks, applying the general principles. Chapter 5
discusses the role disclosure can play in adjusting monetary
policy operational frameworks. Chapter 6 identifies the
strategic choices a central bank faces when considering
climate-proofing its monetary policy operational framework.
NGFS REPORT
12
2.1. Climate change brings new
nancial risks for central banks
Climate change is a source of nancial risk. Climate-
related financial risks arise through two main channels.
Transition risks arise from the significant structural
changes required for economies to adjust towards a
low-carbon economy (disruptive innovations, policy
changes
5
including carbon pricing policies, shifts
in consumer preferences
6
). These transition risks can
lead to assets becoming “stranded”, i.e. losing value
as a result of unanticipated changes in expected cash
flows. Uncertainty surrounding climate change policies
and their pace is one driver of transition risk.
7
Physical
risks arise from the increasing severity and frequency
of extreme climate and weather-related events (e.g.
floods and hurricanes), and chronic shifts in weather
patterns (e.g. temperature increases, rising sea levels).
The materialisation of either risk type can cause
heavy financial losses and impair asset values through
unanticipated changes in their expected cash flows,
8
impacting the creditworthiness of particular issuers, and
giving rise to systemic risk (for more details, see NGFS
2019b, NGFS 2020a). Climate change being an externality,
it may be the case that the associated financial risks are
not sufficiently reflected in prices. Even increased climate-
related disclosure may not result in market prices reflecting
the entire social cost of climate change. The suggestion
instead is to determine this collectively, e.g. through a
political process and the introduction of climate policies.
9
Climate-related nancial risks could have medium to
long-term implications for the economic outlook and
nancial system. Climate change could materially affect
monetary conditions. For instance, abrupt asset price
corrections triggered by climate-related risks may make it
harder for banks or other financial intermediaries to obtain
liquidity in interbank and other short-term funding markets
because of higher perceived counterparty risk or reduced
collateral availability. Falling asset prices also reduce the
value of the collateral available to firms and households to
support credit demand. In the presence of falling asset values,
banks may reduce their credit supply in order to maintain
regulatory capital ratios.
10
Such shocks could alter monetary
policy transmission channels (see Figure 3), and, potentially,
the ability of central banks to safeguard financial stability.
Climate-related nancial risks may damage market
condence, output and nancial stability, and thus
aect both the counterparties and nancial assets
that are used in monetary policy operations. These
risks could impact monetary policy through their effect on
the financial soundness of central banks’ counterparties,
and on the value of assets pledged as collateral or held
outright. If the market values of eligible assets were to
fall excessively, it could reduce the amount of liquidity
available to central bank counterparties. A counterparty’s
access to liquidity could also be curtailed if its exposure
to climate-related risk jeopardises its financial position to
a point where it ceases to meet its central bank’s financial
soundness requirements. Lastly, adverse climate-related
price shocks to assets that are purchased outright may
need to be taken into account when setting quantitative
easing policies and central bank targets.
As sources of nancial risk, climate-related shocks
can generate losses for central banks. While a central
bank’s objective is not to generate profits but to fulfil a
broader mandate, typically related to broader social welfare,
financial losses can nevertheless pose risks to its reputation,
credibility and financial independence and may require
recapitalisation measures.
2. The state of play
5 McGlade and Ekins (2015) estimate that one-third of global oil reserves and half of gas reserves should remain unextracted in order to limit global
warming to 2°C. Stricter national regulations to limit the extraction of petroleum will be necessary in order for countries to achieve their nationally
determined contributions, as pledged under the Paris Agreement.
6 See UN PRI (2019).
7 For example, when policies are introduced gradually, assets may experience a loss in value over time with manageable adjustment costs. For this to
happen, policies must be credible and investors need to understand how to account for them. See Sen, S. and M. T. von Schickfus (2020).
8 See IPCC (2018).
9 For more details see Krogstrup and Oman (2019).
10 See Batten, S., R. Sowerbutts, and M. Tanaka (2016).
NGFS REPORT
13
2.2. Adapting traditional central
bank models to climate change?
Central banks have not yet reached a consensus as to
whether and how their operational frameworks should
incorporate the eects of climate change.
Modern central banking rests on certain commonly
accepted principles. One of them is that, typically, a central
bank does not seek to target individual firms, households,
regions or economic sectors. Another is that, to limit inflation
risks, governments should not have automatic access to
central bank (base) money. These two principles, though
not universally accepted, imply that a central bank should
refrain from using its powers to carry out tasks that do not
fall within its remit or tasks that might more properly be
the responsibility of governments.
11
As far as climate-related risks are concerned,
governments have the principal responsibility for
setting the policy response to climate change and have
a much broader range of tools and policies on hand to
prevent and mitigate it than central banks do. Such tools
may include incentives for agents to shift to low-carbon
activities, perhaps by way of increasing carbon prices via
taxation or the issuance of carbon certificates, supporting
research on and investment in low-emission technologies or
even prohibiting certain activities altogether (Lagarde and
Gaspar 2019, Arezki and Obstfeld 2015, Farid et al. 2016).
If central banks introduce, for example, measures focused
on leading and shaping the financial sector’s response to
climate change, they can complement government-led
action.
Whether and to what extent central banks should
modify their behaviour and approaches in support
of governments’ objectives on climate-related issues
depends, inter alia, on their mandate and on social
norms, which dier across regions. Societal conventions
help shape institutional frameworks such as central
11 See Honohan, P. (2019). For more details, see Tucker, P. (2018).
Figure 3. Monetary policy transmission under climate change strains
Official interest rates
Expectations
Money market
interest rates
Money,
credit
Asset
prices
Exchange
rate
Bank
rates
Wage and
price-setting
Supply and demand in
goods and labour markets
Domestic
prices
Import
prices
Price developments
Feedback loops
and amplications
Feedback loops
and amplifications
Banks’ balance sheet
(market and credit risks)
Changes global economy (GVCs)
and commodity markets
Firms’ solvency & profitability
is hit. Rising NPLs
Changes in risk premia
Stranding of assets
and “green heaven” effect
Scenarios with physical
and transition risks
Higher risk aversion and uncertainty
Changes in preferences
NGFS REPORT
14
bank mandates and therefore influence their room for
manoeuvre in supporting government policies. Thus far,
central banks (or relevant policy committees) with mandates
that explicitly include climate-related objectives are an
exception. Nevertheless, in the NGFS survey mentioned
in the introductory chapter, many central banks indicated
that there is scope in their existing mandates to adjust
their policy frameworks should they decide to cater more
for climate-related challenges.
12
Expectations about central bank actions evolve over
time. As they deployed new instruments to address recent
crises (e.g. the 2007-09 crisis and the fallout of the COVID-19
pandemic), central banks faced increased scrutiny about
their actions and how they manage the side effects without
compromising on their primary objective.
13
The at times
controversial debate surrounding the role that market
neutrality should play in the practical implementation of
monetary policy is a case in point.
Faced with climate-related risks, central banks must
ensure that their operational frameworks remain
ecient for the smooth conduct of monetary policy
within their mandates, while mitigating the risk that their
actions conflict with the broader climate policies needed
to transition to a low-carbon economy. Central banks
should be mindful that their actions can undermine the
transition to a low-carbon economy and consider the
double-materiality perspective of their actions, which
consists of taking into account the effect of climate change
on them, as well as the effects of these actions on climate
change itself.
2.3. Potential courses of action
There is a consensus among NGFS members that, at the
very least, central banks should carefully assess and,
where appropriate adopt, additional risk management
measures to protect their own balance sheets against
the nancial risks brought about by climate change.
As mentioned above, central banks are directly exposed
to climate-related financial risks through their operational
frameworks, and they may incur financial losses if they
fail to protect themselves against those risks. Currently,
central banks’ operational frameworks typically account
for liquidity, market and credit risks through a range of
risk management rules and techniques, which include
financial soundness checks, minimum rating requirements
and other eligibility criteria for collateral, collateral haircuts,
valuation markdowns, due diligence of asset purchases,
and concentration limits. Further work is needed to
determine whether current measures are sufficient or
suitable enough to protect central banks against climate-
related financial risks.
Assessing the appropriate level of protection against
climate-related nancial risks is a challenge for central
banks. These risks are intrinsically difficult to measure
with precision, notably due to the radical uncertainty that
characterises climate risks (tipping points, non-linearities,
regime shifts, etc.), not to mention practical issues such as
data and methodological gaps. As a result, it cannot be
taken for granted that existing risk control measures by
central banks provide adequate protection against climate-
related risks. Central banks need to use appropriate risk
management tools to identify, measure, and, if necessary
address, these risks.
Aside from risk management-driven initiatives, another
reason for central banks to consider action relates to
the potential for adverse consequences that climate-
related shocks could have for the effectiveness of
monetary policy over time. While their materiality is
under investigation at many central banks, it is widely
recognised that climate-related shocks will adversely and
increasingly impact macroeconomic and price stability.
14
These negative impacts may vary depending on the ability
of monetary policymakers to respond and any measures
that are already in place. The recent survey among NGFS
members highlighted that some central banks consider
they are already experiencing some of these effects on the
transmission channels of monetary policy, mainly following
natural disasters.
12 See NGFS (2020b).
13 See Honohan, P. (2019).
14 See NGFS (2020a).
NGFS REPORT
15
15 Indeed, some central banks may decide not to take immediate action, depending on their exposure to climate-related risks and/or due to constraints
such as their mandate and the lack of sufficient research.
Central banks may also consider supporting the
transition to a low-carbon economy using monetary
policy tools, where they have a clear policy remit to
do so. The seriousness of the global climate challenge
suggests that some combination of protective and climate
mitigation approaches may be required, insofar as they
can be balanced with central banks’ existing institutional
objectives. To the extent that the design of monetary policy
instruments may conflict with incentives for a smooth
transition to a low-carbon economy, central banks will have
to assess whether they can adjust their toolkits without
compromising on the efficiency of monetary policy.
However, the main driver for the transition to a low-carbon
economy should remain the action taken and transition
strategy laid out by governments.
Overall, the distinction between “protective” and
“proactive” approaches to climate-related risks is
blurred from an operational viewpoint. Moreover,
central banks need to clarify their climate-related objectives
before designing measures. Some options would allow
central banks to both protect themselves against climate-
related risks and take action to mitigate their effects. Some
climate-related risk protection measures can have positive
side effects for the transition to a low-carbon economy.
Conversely, some proactive measures may give the central
bank’s balance sheet greater protection over the medium
term. However, protective and proactive measures can
sometimes lead to conflicting results. Some proactive
measures may not protect the central bank balance sheet,
and some protective measures may not protect the climate.
Even if the same tools can usually be used to implement
both proactive and protective policies, the scope/calibration
of the polices might be different if used for risk protection
or to promote the transition.
Some considerations may induce central banks to refrain
from adjusting their operational frameworks. In the
short term, these relate notably to operational difficulties
and the risk of miscalibration or of unintended negative
consequences for monetary policy implementation
and for the central bank’s credibility. Climate-related
risk measurement remains a nascent field, and central
banks do not know more than financial markets about
how to measure or price climate-related financial risks.
Depending on the nature of the adjustment, there may
also be constraints on the authority of the central bank.
Taking no action is not viewed as a sustainable option
over time, not least because climate change brings new
nancial risks for the central bank. Making adjustments
prematurely, without suitable knowledge, data, or legal
clarification regarding the central bank’s mandate may
undermine its credibility.
15
That said, the scientific consensus
is that the damage associated with unmitigated climate
change will be high and increasing over time, and that the
risk of catastrophic tail events is by no means negligible.
Such significant economic damage could force central banks
to adjust their operations in a precipitous way, hence the
need for central banks to at least consider the issue now.
NGFS REPORT
16
Assessing potential climate-related adjustments to
monetary policy operations, in general, is challenging
because of the heterogeneity of central bank operational
frameworks. Most monetary policymakers focus on price
stability as a primary objective,
16
which typically means
low and stable inflation and/or exchange rate stability.
However, even monetary policymakers with similar primary
objectives may implement their policies differently.
Indeed, the operational frameworks of central banks can
vary significantly in terms of their operational targets, the
liquidity environment in which they operate, and the choice
of preferred instruments.
Regardless of these dierences, potential adjustments
can be assessed against four general principles. These
are: (1) Consequences for monetary policy effectiveness;
(2) Contributions to mitigating climate change; (3) Effectiveness
as risk protection measures; and (4) Operational feasibility.
3.1. Consequences for monetary
policy eectiveness
While climate-related risk adjustments may be helpful in
terms of risk identification and mitigation, some may have
negative consequences for the conduct and effectiveness
of monetary policy operations, which would likely count
against their adoption. Monetary policy operations are
often designed to minimise intervention in financial markets
while maximising the pass-through of policy measures and
treating economic agents equally and fairly. Climate-related
adjustments to the operational framework which result in
constraints on a central bank’s policy space, or which strongly
disincentivise participation in monetary policy operations
or reduce the effective transmission of monetary policy are
unlikely to be considered desirable. Such effects may also
arise if these constraints have not yet manifested but agents
are expecting them. This assessment should evaluate the
extent to which any such measures conflict with monetary
policy transition mechanisms.
3.2. Contributions to mitigating
climate change
Climate-related adjustments to monetary policy operational
frameworks should be assessed in terms of their relevance
and ability to mitigate climate-related risks and/or support
the transition to a low-carbon economy. Adjustments should
be assessed in terms of whether they will conflict with, delay,
support, or be conducive to a smooth transition. In practice,
assessing the effectiveness of any measure on mitigating
the impact of climate change should rest on the principle of
proportionality that many central banks follow, according
to which any potential side effect of the measures should
be weighed against its benefits.
3.3. Eectiveness as risk protection
measures
Central bank risk management frameworks typically aim
to ensure that monetary policy objectives can be achieved
with the lowest financial risk possible. Changes to these
frameworks to take climate-related risks into account should,
in principle, improve the identification, measurement and
mitigation of financial risks. This assessment should consider
whether climate-related adjustments would improve or
impair a central bank’s financial risk management. Central
banks should also be mindful of mitigating excessive asset
price adjustments stemming from their risk management
framework.
3.4. Operational feasibility
Climate-related adjustments to central bank operational
frameworks require (i) access to sufficiently robust and
broad-based climate-related risk data; (ii) expertise in
climate-related financial risk management; and (iii) sound
methodologies and models to embed climate-related
measures into operational frameworks.
16 See NGFS (2020b).
3. Principles for assessing potential climate-related adjustments
to monetary policy operational frameworks
NGFS REPORT
17
Depending on the course of action chosen, central banks
may also assign different weights to these principles.
For instance, a central bank that is concerned
about climate-related risks predominantly from
a financial risk management perspective is more
likely to assign a higher weight to a measure’s
effectiveness for risk protection than to its ability
to support the transition to a low-carbon economy.
Organisational factors may also influence the relative
importance of the principles. Central banks with limited
resources may give comparatively more importance to
operational feasibility.
Box 1
What is the carbon performance of monetary policy operations?
Central banks, like other institutions, face increasing
demands for greater transparency on their carbon
performance. An increasing number of businesses
and financial institutions are assessing their carbon
performance, either voluntarily or to meet requirements
set by law. Given the prominent role central banks play
in the financial system, the impact of monetary policy on
climate change has been subject to increased scrutiny in
some jurisdictions.
Assessing the carbon performance of monetary policy
operational frameworks is particularly relevant for
two core monetary policy operations: asset purchases
and collateral policies. As of today, few central banks
have assessed and published their carbon performance.
The Bank of England published its first climate-related
financial disclosure in 2020, following Task-force on Climate-
related Financial Disclosures (TCFD) recommendations.
The Riksbank has announced its intention to report on the
carbon footprint linked to its corporate bond purchase
portfolio in the first half of 2021.
A consistent and comprehensive assessment of the
CO
2
equivalent (CO
2
e)
1
footprint of monetary policy
operations is challenging. This is due to data coverage
problems as well as methodological issues that are not
specific to central banks, but are likely to be of a larger
order of magnitude given the scope, scale and specificities
of their operations.
•
At the current juncture, assessing the CO
2
e performance
of some asset classes that are relevant for monetary
policy is difficult. For instance, assessing covered bonds
and securitisation products, which are accepted by
some central banks as collateral and/or held in policy
portfolios, would likely require a look-through approach
on their underlying assets. However, detailed data on
the carbon performance of underlying bank loans
are typically unavailable or only to a limited extent.
Initiatives are ongoing to find solutions, but there is as
yet no commonly agreed approach for these types of
assets. Similarly, CO
2
e emissions of collateral consisting
of small and medium-sized enterprise credit claims
are not currently disclosed, which requires the central
bank to use rough proxies (e.g. to apply an average
estimate based on economic sector level data or apply
a de minimis rule, etc.).
•
More fundamentally, CO
2
e accounting issues, such
as double counting, imported emissions and indirect
emissions, are all the more significant for those
central banks that, in order to facilitate the smooth
implementation of monetary policy, allow monetary
policy counterparties to pledge a very wide variety of
asset classes as collateral and that hold diverse asset
portfolios.
•
While central banks have control over the eligibility
criteria applied to asset purchases and collateral, there
is a key difference between the carbon footprints of
these two types of assets. The composition of the
collateral pool is dynamic and to a large extent beyond
the control of the central bank, meaning their carbon
footprint may be more variable. By contrast, monetary
policy outright purchases imply a more direct and
often longer-term exposure to the climate-related
risks associated with the issuer.
1 CO
2
e means carbon dioxide equivalent, which is used to compare emissions of various GHG on the basis of their global warming potential (GWP)
by converting amounts of other gases into the equivalent amount of carbon dioxide with the same global warming potential.
NGFS REPORT
18
4.1. Identifying the options
The menu of options that a central bank could consider in
order to factor climate-related risks into its operational
framework is, in theory, large. Adjustments could be
considered across all the main operational functions that
central banks carry out for the purposes of implementing
monetary policy: credit operations, collateral and asset
purchases.
Nine stylised options are reviewed in the report, covering
the three main policy fields outlined above (see Table 3
for the list and descriptions). These have been selected
because they are relevant to multiple central banks and
relate to existing tools. Some options represent a greater
departure from standard central bank policies than others.
Other options for adjusting central banks’ operational
frameworks are conceivable, and these could be evaluated
using the assessment framework detailed in Chapter 3.
The report focuses on potential measures on the asset side
of a central bank’s balance sheet. It does not look into the
liability side of the balance sheet. Hence, the stylised options
listed in Table 1 pertain to liquidity-providing instruments, while
liquidity-absorbing instruments, i.e. reserve requirements,
term deposits, issuance of central bank bills and the like, are
not discussed in any detail. However, it is acknowledged that
such instruments might also be relevant for central banks,
Table 3. Selected stylised options for adjusting operational frameworks to climate-related risks
Credit operations
a
(1) Adjust pricing to reflect
counterparties’ climate-related
lending
Make the interest rate for central bank lending facilities conditional on the extent to which a
counterparty’s lending (relative to a relevant benchmark) is contributing to climate change mitigation
and/or the extent to which they are decarbonising their business model.
(2) Adjust pricing to reflect the
composition of pledged
collateral
Charge a lower (or higher) interest rate to counterparties that pledge a higher proportion of low-carbon
(or carbon-intensive) assets as collateral or set up a credit facility (potentially at concessional rates)
accessible only against low-carbon assets.
(3) Adjust counterparties’ eligibility Make access to (some) lending facilities conditional on a counterparty’s disclosure of climate-related
information or on its carbon-intensive/low-carbon/green investments.
Collateral
b
(4) Adjust haircuts
c
Adjust haircuts to better account for climate-related risks. Haircuts could also be calibrated such that
they go beyond what might be required from a purely risk mitigation perspective in order to incentivise
the market for sustainable assets.
(5) Negative screening Exclude otherwise eligible collateral assets, based on their issuer-level climate-related risk profile for
debt securities or on the analysis of the carbon performance of underlying assets for pledged pools
of loans or securitised products. This could be done in different ways, including adjusting eligibility
requirements, tightening risk tolerance, introducing tighter or specific mobilisation rules, etc.
(6) Positive screening Accept sustainable collateral so as to incentivise banks to lend or capital markets to fund projects and
assets that support environmentally-friendly activities (e.g. green bonds or Sustainability Development
Goals-linked assets). This could be done in different ways, including adjusting eligibility requirements,
increasing risk tolerance on a limited scale, relaxing some mobilisation rules, etc.
(7) Align collateral pools with
a climate-related objective
Require counterparties to pledge collateral such that it complies with a climate-related metric at an
aggregate pool level.
Asset purchases
d
(8) Tilt purchases Skew asset purchases according to climate-related risks and/or criteria applied at the issuer or asset level.
(9) Negative screening Exclude some assets or issuers from purchases if they fail to meet climate-related criteria.
a Credit operations are widely used to provide aggregate liquidity and usually take the form of collateralised lending.
b Collateral policy denes the range of assets that can be pledged to secure central bank credit operations, as well as the risk control measures that apply
to them.
c Annex 1 expands upon the dierent approaches for haircuts and valuation adjustments.
d Central banks may buy a variety of assets from both public and private sectors, typically in an eort to exert greater inuence on longer-term interest rate
levels and spreads while improving market liquidity.
4. Reviewing potential options
NGFS REPORT
19
depending on the liquidity environment in their respective
jurisdictions. This may provide a starting point for further
analyses by the NGFS on this specific field in the future.
This review is without prejudice to each NGFS central
bank’s mandate, legal environment and individual
assessment. It is based on the criteria defined in Chapter 3
and leverages on the practical expertise and experience
of NGFS central banks. These assessments are not
recommendations, nor are they indicative of members’
preferences. Rather, they provide a first collective attempt to
compare the considerations which may pertain to different
options in a structured and consistent manner.
4.2. Summary assessment
Since monetary policy operational frameworks vary
significantly across jurisdictions, this assessment
necessarily remains generic, i.e. it is only described at
a high level. Strategic choices that cut across the various
options (such as metric specification, data availability, risk
assessment and risk tolerance) are discussed in Chapter 6.
Adjusting central bank operational frameworks to reect
climate-related considerations more fully is potentially
feasible, provided additional central bank-specic work
is done. Yet accounting for climate-related risks in central
bank operations has to overcome a range of practical and
analytical challenges, including data gaps and the difficulties
associated with measuring climate-related risks. There is a
priori no “one size fits all” option that clearly maximises all four
criteria identified in Chapter 3, and central banks face trade-offs
when integrating climate-related risks into their operations
(see Chapter 6). The assessment below summarises how the
selected generic options perform against various criteria.
The detailed assessment including case studies from central
banks already pioneering the use of specific measures in their
operational frameworks can be found in Annex 1.
Consequences for monetary policy
eectiveness
Assessing the implications of these options for
the eectiveness of monetary policy operations is
challenging. While the impact on the flexibility or scope
of policy operations can be evaluated in light of practical
experience, predicting the broader consequences for the
effectiveness of monetary policy transmission is more
difficult. The latter is a function of each central bank’s actual
operational framework and the financial ecosystem in which
it operates. Further jurisdiction-specific work is needed
on this aspect, but a few preliminary points can be made.
Several options may run the risk of curtailing, more or
less signicantly, the scope for central bank operations
and the policy space. This risk may be more significant in
options that aim to (i) exclude counterparties from credit
operations on the basis of their carbon footprint or carbon
disclosure; and (ii) screen out assets potentially representing
a significant share of the purchasable universe or of eligible
collateral. These options can offer a pragmatic way for
central banks to mitigate tail risks. Their impact on the
monetary policy space depends on their design and how
stringent they are.
Options which may materially reduce the available
monetary policy space, or which are substantively
prejudicial to its ecacy, should be considered and
designed cautiously. One key point of vigilance concerns
the potential unintended consequences that some options
may have for financial stability. When designing concrete
measures, central banks need to retain flexibility as regards
implementation in order to prevent unintended impacts on
their ability to achieve their monetary policy objectives or
on financial stability. When implementing climate-related
options, central banks need to reserve the right to take
account of prevailing financial market conditions, especially
in times of market stress, and possibly define “escape clauses”.
The implications for the eectiveness of monetary
policy of some options may be negligible or still require
further assessment in the light of each central bank’s
circumstances. This is the case for those options that
focus on (i) conditioning the pricing of credit operations
on a low-carbon lending benchmark; (ii) adjusting haircuts
and valuations to account for climate risks; (iii) aligning
collateral pools; and (iv) tilting asset purchases.
Contribution to mitigating climate change
A few options may be more impactful from a climate-
related perspective than others. These include
measures aimed at (i) adjusting the pricing of targeted
credit operations to a climate-related lending benchmark;
(ii) positively screening collateral; (iii) aligning collateral
NGFS REPORT
20
pools; and (iv) tilting asset purchases. These measures
would typically leverage and foster market mechanisms.
They typically consist of modifying existing tools without fully
overhauling their design (e.g. leveraging pricing schemes for
targeted credit operations) in order to encourage lenders to
originate or invest more in low-carbon and transition assets.
Seen from this perspective, they could be consistent with
the smooth implementation of monetary policy, although
still technically challenging to operationalise.
Eectiveness for risk protection purposes
Many of the options reviewed would probably better shield
central bank balance sheets against increasing nancial
risks, notably those options that would directly reduce risk
exposure (to issuers or counterparties). The following options
are viewed as being probably risk-protective: (i) adjusting
eligible counterparties; (ii) adjusting haircuts and valuations;
(iii) negatively screening collateral; (iv) aligning collateral pools;
(v) tilting asset purchases; and (vi) negatively screening asset
purchases. However, for some of them, the potentially positive
impact may be diluted, or in some cases even outweighed, if
the reduction in the eligible universe leads to higher financial
risk concentrations. A priori and contingent on each central
bank’s mandate, options designed from a financial risk
perspective (e.g. negative screening) may be less exposed
to legal risks and challenges than others designed to support
climate-related objectives, especially if the latter are seen as
subsidising some economic sectors, issuers or assets.
For many options, the actual impact from a risk protection
perspective is difficult to assess without a more
detailed specification. This is the case for the options
consisting of adjusting the pricing of targeted credit operations
(i) to a climate-related lending benchmark; or (ii) to pledged
collateral. Risk protection and the contribution to mitigating
climate change may, in some cases, converge as some
operational options serve both purposes (see Annex 1).
Operational feasibility
All options entail signicant changes to central bank
operational frameworks. Factoring in climate-related
considerations may imply procuring additional specialist climate
data, adapting IT and reporting systems, revising internal
processes, and rewriting operational terms and conditions.
The least challenging options to operationalise are the
least sophisticated ones in terms of addressing climate-
related risks, but potentially costlier in terms of monetary
policy effectiveness. These include (i) adjusting eligibility
criteria for counterparties; (ii) screening collateral; and
(iii) screening purchasable assets.
Conversely, the options that are less likely to entail adverse
consequences for monetary policy effectiveness are typically
associated with somewhat higher operational complexity.
This is the case for (i) adjusting the pricing of targeted
credit operations to a lending benchmark; (ii) haircut and
valuation adjustments; (iii) aligning collateral pools; and
(iv) asset purchase tilting. Whether any additional complexity
would be warranted to achieve associated reductions in
financial risk or improved climate outcomes would need
to be assessed on a case-by-case basis.
Table 4. Simplied comparative assessment of the selected generic options under review
ADJUSTING
PRICING TO
LENDING
BENCHMARK
ADJUSTING
PRICING TO
COLLATERAL
ADJUSTING
COUNTERPARTIES’
ELIGIBILITY
HAIRCUT
ADJUSTMENT
NEGATIVE
SCREENING
POSITIVE
SCREENING
ALIGNING
COLLATERAL
POOLS
TILTING NEGATIVE
SCREENING
CONSEQUENCES FOR
MONETARY POLICY
EFFECTIVENESS
CREDIT OPERATIONS
COLLATERAL
ASSET PURCHASES
CONTRIBUTION TO
MITIGATING CLIMATE
CHANGE
EFFECTIVENESS AS RISK
PROTECTION MEASURE
OPERATIONAL
FEASIBILITY
POTENTIAL IMPACT :
STRONGLY POSITIVE MINIMAL STRONGLY NEGATIVE
POSITIVE NEGATIVE
(1) (2) (3) (4) (5) (6) (7) (8) (9)
The assessment is based on qualitative expert judgement, and more formal quantitative analysis may be needed. It aims to guide the reader through
the report and should not be interpreted as recommending any measure. Colour-coding is used to avoid any “netting” across criteria. The table uses
a limited number of colours for reasons of simplicity. More nuanced analyses of options are provided in Annex 1.
NGFS REPORT
21
4.3. Open questions
This review highlights two critical questions that central banks
need to address. One relates to the sequencing of potential
actions. The other to the consistency between them.
Sequencing
Some options are more challenging than others and
require signicant lead times and preparatory work.
Prioritisation of the options needs to be informed by data.
Many central banks may deem it important to improve their
access to relevant specialist climate data and their ability
to measure climate-specific risks on both a backward and
forward-looking basis. While much progress has already
been made on relevant metrics, central banks need to
embed the use of these metrics both internally and in the
financial systems for which they are responsible.
More generally, central banks need to assess whether
to adopt a “learning by doing approach” or to design a
comprehensive climate-adjusted framework. The choice
between the two depends on their mandate, their financial
environment, and the guiding principles underpinning
their operational framework. The first approach is likely
to imply gradual step-by-step adjustments. It echoes
practices at several central banks in the introduction of
new monetary policy instruments during recent crises. It can
consist of selectively adapting the operational framework,
focusing first on those issuers and/or assets for which
data and methodologies are more mature and robust.
The second approach is likely to imply longer lead times
and/or acknowledging that data and methodologies need
to be relied on which may be comparatively less robust for
some assets or issuers. In practice, it can consist of adapting
the operational framework to the greatest extent, which
may entail giving preference to a prudent calibration of
the measure from a climate perspective.
Consistency
Central banks have to be mindful of the consistency of
their operational frameworks so as to avoid creating
unwarranted incentives or unintended consequences, or
introducing bias into market functioning. Policy design
and the sequencing of climate-related options will likewise
need to respect this general principle.
This implies that central banks need to carefully
assess whether the climate-related measures they
opt for are mutually consistent and/or interdependent.
For instance, they should clarify whether requiring
monetary policy counterparties to make climate-related
disclosures can be considered independently of other
climate-related measures (e.g. in terms of counterparty
ineligibility or in terms of credit operations price
adjustment). Similarly, depending on how their operational
framework is designed, central banks need to assess
whether it is consistent to adjust their collateral policy
to climate-related risks without considering doing the
same for purchases if they both cover similar assets (and
vice versa).
NGFS REPORT
22
Enhanced disclosure of climate-relevant data is
an operational and policy issue that cuts across
the potential menu of central bank policy options.
Disclosure requirements may be imposed by central banks
or other public authorities depending on their respective
responsibility within their jurisdiction. They could help
foster the greater availability of climate-related data
and improve the measurement of climate-related risks
across the financial system. Data availability seems to be a
material constraint to assessing climate-related exposures.
17
Such requirements could be implemented in multiple ways
and are not confined to any specific operational framework.
In broad terms, disclosure requirements can take
two forms, depending on whether the beneficiary of
the enhanced body of information is public or internal.
Where this falls within their power, central banks could
impose public disclosure requirements on other parties
by leveraging commonly agreed or regulatory reporting
frameworks. Information disclosed within these frameworks
would benefit not only the central bank, but also market
participants, other public authorities and economic agents
at large. Alternatively, central banks could impose reporting
requirements on other parties but keep the data confidential
and use it only for their internal assessments. In this case,
the enhanced body of information would benefit only the
central bank.
Whether climate data should be disclosed only to
the central bank or more broadly will depend on the
objective of disclosure. If data are collected purely for
risk-protective purposes, theoretically they might not
need to be made public. But greater disclosure is likely to
be more helpful in terms of enabling the market to price
climate risk more fully and more consistently with the aim
of supporting climate transition.
Central banks should also assess whether or not to
disclose climate-related information on their own
policy operations and nancial activities. This could
be motivated by considerations about transparency and
accountability to the public about the climate-related
risks and carbon footprint of their operations. It could also
serve to signal a central bank’s commitment to enhancing
climate risk information.
Central banks can play a key role in the development of
data, metrics and best practices – and are well positioned
to do so, given their experience in setting reporting
standards for statistical, supervisory and regulatory
purposes. Establishing best practices in the field of climate-
related data is likely to be an iterative process.
5.1. Is disclosure a prerequisite for
other potential adjustments?
Increased disclosure of climate data is critical for helping
central banks to better understand their exposures to
climate-related risks. Enhancing disclosures can reduce
information asymmetries, contribute to research and
analysis on the impact of climate-related risks on monetary
policy transmission and operations, and support a better
measurement of central banks’ exposures to climate-related
risks. Over the medium term, regular and standardised
disclosures could contribute indirectly to protecting the
monetary policy transmission mechanism by reducing
potential informational market failures (see Box 2).
Increased availability of climate-related information
may, in some cases, be a prerequisite for adjusting
operational frameworks. The recalibration of some of
the measures explored in Chapter 4 and Annex 1 may
require more extensive and robust information than is
currently available.
Nevertheless, some climate-related adjustments
to operational frameworks could be developed in
parallel with data disclosure and reporting initiatives.
One adjustment that could potentially be made without
additional data is to accommodate climate-related financial
innovations (e.g. the development of sustainability-linked
bonds) within the eligible collateral universe, where feasible.
Other adjustments that could be developed alongside
17 See, for example, Financial Stability Board (2020).
5. Disclosure
NGFS REPORT
23
Box 2
To what extent does climate-related disclosure
aect the pricing of assets (notably xed income assets)?
Disclosing climate change-related metrics may
provide investors with important information on the
management of climate-related risks. At the same time,
pricing climate-related risks may be difficult if information
is not available in a comparable format. In recent years,
initiatives to develop and promote harmonised climate-
related disclosures have gained momentum. On the
one hand, many large companies around the globe are
following the recommendations of the Financial Stability
Board’s Task Force on Climate-related Financial Disclosures
(TCFD), albeit with varying degrees of consistency and
completeness. On the other, private market data providers
have developed what are known as environmental, social
and governance (ESG) scores that seek to consolidate
quantitative and qualitative environmental information
and enhance its comparability.
Although academic studies may at rst glance appear
to paint a mixed picture of the extent to which climate
disclosures impact asset pricing, a granular approach
does reveal some patterns.
1
For example, investors in
certain energy-intensive sectors (such as mining, cement
production or the oil industry) may be more sensitive
to climate disclosures by issuers. Chart 1 illustrates the
different means of cumulated changes in firm valuations
since the Paris Agreement, depending on whether the
firm in question reported its carbon emissions according
to Thomson Reuters or Bloomberg. The differences are
in line with the results of some recent studies which find
that environmental disclosures matter for profitability
expectations and reduce information asymmetries at
least in the most climate relevant activities, such as
the energy sector.
2
Further analysis of the data shows
that the risk perception related to a firm not disclosing
climate change data may have increased over time, as
the difference is stronger for the 2017-19 sub-period.
Lastly, looking at different financial market instruments,
equity and bond market investors could also respond
differently to climate disclosures, depending on whether
they expect non-disclosure to signal lower future cash
flows or whether it is also thought to increase short to
medium-term credit risk.
1 See, for example, Friede, G., T. Busch, and A. Bassen (2015).
2 See, for example, Bolton, P. and M. Kacperczyk (2020); Downar, B., J. Ernstberger, S. Reichelstein, S. Schwenen, and A. Zaklan (2020); Hsu, P.-H., K. Li,
and C.-Y. Tsou (2018); Alessi, L., E. Ossola, and R. Panzica (2019); and Bui, B., O. Moses, and M. N. Houqe (2019).
…/…
improvements in data disclosure are those designed for
specific sub-categories of operations (e.g. a segment of a
purchase programme, a sub-category of collateral), where
a central bank deems data quality and availability to be
sufficient. Central banks may also opt for workaround
solutions (e.g. using sector averages rather than asset or
issuer-specific data).
Central banks should aim to strike a balance between
waiting for robust and comprehensive data to become
available and the risks associated with delaying action.
The cost of acting only once robust datasets and methods
are established could be much higher than the cost of
acting now given limited information. They should also
consider that a dynamic relationship might exist between
central bank measures (even if only partial or gradual) and
stakeholder incentives to disclose and produce better
data and methodologies. However, for policies based on
disclosures to be efficient, it is critical that the data to be
disclosed are reliable and comparable. Introducing policies
before robust datasets and methodologies are available
could result in policy measures being inefficient.
NGFS REPORT
24
Chart 1. Mean stock price change is higher for rms disclosing carbon emissions in the energy and industrial sectors
Energy (322 equity ISINs) Industrial (1,460 equity ISINs) Utilities (261 equity ISINs)
0 .2 .4 .6 .8 1
Density
−1 0 1 2 3 4
Return from 2015 to 2019
no GHG reporting in Thomson Reuters or Bloomberg
GHG reporting
kernel = epanechnikov
Energy sector
0 .2 .4 .6 .8
Density
−1 0 1 2 3 4
Return from 2015 to 2019
no GHG reporting in Thomson Reuters or Bloomberg
GHG reporting
kernel = epanechnikov
Industrial sectors
Utilities sector
0 .2 .4 .6 .8
Density
−1 0 1 2 3 4
Return from 2015 to 2019
no GHG reporting in Thomson Reuters or Bloomberg
GHG reporting
kernel = epanechnikov
Sources: Bloomberg and Thomson Reuters, own calculations.
Notes: Sample comprises 2,043 equity instruments of listed non-nancial rms between end-2015 (start of the Paris Agreement) and end-2019. The disclosure
measure is dened as a dummy variable that takes the value of one if either Bloomberg or Thomson Reuters provide emission data for the respective rm in 2018,
and zero otherwise. Mean comparison tests suggest a non-signicant dierence in stock price changes in the utilities sector, but signicant dierences at the 1%
condence level for rms in the energy sector and the industrial sector.
Recently, awareness of the potential impact of climate
change on credit risks and therefore its importance
for xed income investors has grown.
3
Disclosure of
climate metrics related to issuers may be of importance
for an investor’s credit risk assessment.
4
Also, investment
rules such as exclusion criteria may reduce demand for
bonds of issuers that do not disclose climate-related
information.
Looking at a snapshot of global investment grade
rated corporate bonds suggests that environmental
disclosure has a signicant and negative impact on
bond yields (see Chart 2). The magnitude of this impact
is small, however, when compared to the more traditional
factors related to term and default premia. Environmental
disclosures by issuers, as measured by a positive value of
the environmental sub-index of the ESG disclosure score
calculated by Bloomberg, decrease bond yields and thus
increase bond valuations, in particular in the industrial
and energy sectors. This result would suggest that, in line
with the patterns observed in stock market studies, bond
investors attach the highest premium to transparency on
ESG issues in sectors where climate risk may matter most
for default probabilities.
3 See, for example, IMF (2019).
4 It should be noted that this analysis does not concern what are known as green bonds, which do not imply a different credit risk than other bonds
of the same issuer.
Chart 2. Impact of selected variables on bond yields
(bonds issued in euro; in percentage points)
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
Downgrade premium Term premium (by year)
Premium (%)
Disclosure premium
UtilitiesIndustrialEnergy
Sources: Bloomberg, own calculations.
Notes: Downgrade premium refers to the impact of a credit rating
downgrade on a bond’s mid-yield to maturity and is used for a proxy
for the default premium. The disclosure factor is a dummy variable that
takes the value of one if the issuer rm disclosed environmental
information as part of its ESG disclosure score provided by Bloomberg
in 2017 and/or 2018, and zero otherwise. The chart reects the results
of a cross-section regression of 277 bonds denominated in euro,
issued between 2000 and 2019.
The disclosure premium coecients were signicant at the 1% condence
level for the energy and industrial sectors, whereas the coecient for
utilities was not signicant. The results for bonds issued in US dollar were
qualitatively similar but less signicant for the energy and industrial sectors,
whereas the disclosure coecient for the utilities sector was positive.
However, the sample for the latter sector in US dollar is very small and
therefore not representative. The results remained qualitatively
unchanged when the models were tested on out-of-sample data.
NGFS REPORT
25
5.2. Requiring disclosure from eligible
collateral issuers and/or monetary
policy counterparties
Introducing disclosure requirements in monetary policy
operations could help foster harmonised, transparent,
reliable and comparable data. This could be done as
part of collateral policy, credit operations, counterparty
policy, and/or asset purchase programmes. Disclosure
requirements could be imposed either directly on monetary
policy counterparties (banks, asset issuers) or indirectly by
requiring additional information from counterparties about
the underlying assets of securities pledged as collateral (e.g.
data on the energy efficiency of buildings for the residential
mortgage loans underlying certain securitisation products).
Disclosure requirements can come in various shades.
They can serve to enhance the central bank’s risk
management or promote market transparency in order
to address informational market failures and foster climate
transition.
•
Disclosure could be mandatory and serve to
discriminate between counterparties, assets or issuers
that do not meet prescribed eligibility criteria. Such
criteria may be more or less stringent depending on
their design (full vs. partial disclosure, with or without a
phasing-in period). While powerful, a stricter disclosure
approach may potentially conflict with the smooth
implementation of monetary policy, if it translates into
an excessive exclusion of assets available for monetary
policy operations or counterparties.
•
Alternatively, disclosure can be designed more as
an incentive. Monetary policy stakeholders might be
encouraged to self-report their climate-related risk
exposures, for instance by means of a price incentive
in targeted credit operations, a haircut adjustment, or
the tilting of asset purchases. For structured finance
and non-marketable assets, compliance with a climate-
related loan-level reporting standard could be required,
incentivising financial institutions to obtain relevant
information from their customers. This could increase the
information available on segments where it is currently
most limited, such as small and medium-sized enterprises
(SMEs) and households. Disclosure requirements may be
best suited for risk identification purposes if introduced
at the counterparty or collateral level.
Climate-related reporting requirements would come
at a cost for reporting agents. Central banks should
carefully define their objectives and needs for disclosure
(e.g. disclosure by all or only some issuers, assets, or
counterparties) as well as any additional verification needs
(e.g. whether existing or new requirements are used, what
metrics will be calculated, how data will be collected, and
who will monitor the process). An adverse selection problem
may occur, whereby more-polluting firms may be afforded
better treatment due to their more developed disclosures. To
avoid placing smaller entities at a disadvantage, simplified
requirements may need to be implemented for entities
below a given size threshold.
To reduce the operational burden of disclosure
requirements and issues associated with comparability
and transparency, central banks should seek to leverage
existing reporting frameworks, building on Task Force
on Climate-related Financial Disclosures (TCFD) and
forthcoming International Financial Reporting Standards
(IFRS) requirements. Alternatively, central banks could
refer to available legal frameworks (e.g. the Non-Financial
Reporting Directive (NFRD) in the European Union (EU)).
5.3. Disclosing the central bank’s own
exposures to climate-related risks
Central banks may wish to disclose information on their
own exposures to climate-related risks and on their
climate strategy and performance. This would help set a
positive example to assist market participants in developing
their own disclosure frameworks and incorporating climate-
related considerations into their investment and lending
decisions (see Box 3).
NGFS REPORT
26
Box 3
Case study: Bank of England – approach to TCFD disclosures
In 2020, the Bank of England published its first climate-
related financial disclosure, setting out its approach to
managing the risks from climate change across its entire
operations, and the steps taken to improve the central
bank’s understanding of these risks.
Following the TCFD framework, the report covered the
Bank of England’s:
•
governance structures and processes used to manage
climate-related risks;
•
approach to setting climate strategy and managing
its implementation;
•
approach to climate-related financial risk management,
including the metrics used to measure and monitor
climate-related risks and the climate targets on the
central bank’s physical operations.
A key section of the disclosure assessed the financial
risks from climate change in the central bank’s financial
asset portfolios, including its portfolios held for policy
purposes. This used a range of metrics covering carbon
footprint and exposures to physical and transition risks.
By far the largest proportion (96%) of assets held in the
Bank of England’s financial asset portfolios is held in a
separate legal vehicle, to implement the Monetary Policy
Committee’s (MPC) asset purchase programme. Of this,
based on end-February 2020 data, 98% is invested in
United Kingdom (UK) sovereign government bonds and
2% in UK sterling corporate bonds. The climate metrics
of the Bank’s portfolio are therefore materially aligned
with that of the UK.
In terms of the carbon footprint, because the UK has the
second-lowest carbon intensity in the G7, the portfolio’s
carbon footprint was lower than a comparable reference
index. For the corporate bond portfolio, some more
experimental metrics were used in an attempt to assess
alignment with international climate targets. These metrics
showed that a gap remains between the associated carbon
outputs of those holdings and Paris Agreement goals, in line
with the economic adjustments required across the economy
to meet the UK’s goal of net zero emissions by 2050.
Transition risk was assessed by examining portfolio
exposures to fossil fuel-related activities, and physical
risk was assessed using specialist climate modelling that
seeks to identify geographical vulnerabilities. Estimates
suggested that aggregate transition and physical risks
to the portfolios are modest in the short term – but the
Bank of England recognises that there is scope to take
this analysis further and is committed to developing its
approach as methodologies improve.
This exercise was undertaken with external data providers,
as part of looking across a range of metrics. The approach
enabled the central bank to cross-check physical and
transition risk measures from different sources and
helped to advance thinking on the assessment and
management of climate-related financial risks. The process
also identified a number of current limitations on data,
analysis techniques and modelling methodologies.
Some of the lessons learned include:
1. The need to manage some gaps in emissions data.
Data for some smaller unlisted corporates can be
limited and the reporting of scope 3 data by companies
can be inconsistent.
2. The need to look carefully at how metrics are constructed.
For example, forward-looking temperature alignment
metrics are, in principle, an important analytical tool.
However, differences between alternative modelling
approaches and the lack of comparability among
them can lead to quite different estimates and make
it difficult to draw conclusions from them.
3. There are many potential ways to assess transition
risk. The report focused on a simple and relatively
straightforward measure – exposures to fossil
fuel-related activities – but more advanced scenario
analysis approaches are a possibility.
4. The outputs of physical risk modelling, in particular,
can be difficult to validate independently due to the
specialist nature of these climate models.
The Bank of England is committed to further enhancing
its approach and hopes that its report will help others’
engagement with this type of analysis.
NGFS REPORT
27
Looking beyond the specic adjustment options for
monetary policy operational frameworks when dealing
with climate change, central banks are faced with other
overarching strategic issues, which are to a large extent
interrelated with one another. Developing a stance on
these interrelated strategic issues is a precondition for a
proper evaluation of which adjustments are best suited
to each central bank’s mandate, legal environment and
preferences. Figure 4 maps out, in generic terms, some of
the strategic questions that arise throughout the process
of incorporating climate-related risks into monetary policy
operational frameworks.
Figure 4. Strategic choices for adapting monetary policy operational frameworks to climate-related risks
STRATEGY
RISK MANAGEMENT
VS OTHER
FUNCTIONS
METRICS & DATA
MEASURES
HOW ARE CLIMATE CHANGE RISKS DEFINED?
WHAT IS CENTRAL BANKS’ TOLERANCE TOWARDS THEM?
TRADE-OFF: ACCURATE RISK ASSESSMENT VS SWIFT IMPLEMENTATION
EFFECTIVE MONETARY POLICY IMPLEMENTATION?
BACKWARD OR FORWARD-LOOKING METRICS?
NEED FOR DATA GATHERING
HOW ARE RISK MITIGATION MEASURES DESIGNED ?
6.1. Risk tolerance and assessment
Central banks could benet from developing a clear
strategic view of their appetite for climate-related risks,
which will depend on their own risk tolerance and that of
their institutional environment. Introducing risk-mitigating
measures inevitably impacts the design of monetary
instruments and their implementation. A central bank’s
risk tolerance determines to a large extent its assessment
of the prerequisites that must be in place if it adjusts its
operational framework and if it is willing to deviate from
market standard practices. When weighing the importance
of climate-related risks, central banks must be mindful of
the impact that changes to their risk tolerance may have on
the effectiveness of monetary policy (see Chapters 3 and 4).
Central banks can develop in-house climate-related
risk assessment frameworks. This can help fully integrate
climate-related risk into their standard credit risk frameworks.
One benefit of an internal assessment framework is that it
can cover the exact universe of assets and issuers eligible
as collateral or for asset purchases and match the horizon
of each central bank’s exposures. Developing an internal
credit assessment system can, however, be challenging and
burdensome. In the longer run, such an approach could
complement credit rating agencies’ scores.
Central banks should decide on how forward-looking they
wish their risk mitigation frameworks to be. Traditional
risk models are based on historical market pricing data.
This approach, while used for standard financial risks, is
ill-suited to modelling future climate-related risks, given the
very nature of climate change. Historical data are indeed very
unlikely to be a good guide to the future (TCFD, 2017). This
calls for using scenario-based approaches to climate-related
risks. Yet it is challenging to model the financial impacts of
6. Strategic choices when dealing with climate change
NGFS REPORT
28
climate-related risks across a range of plausible climate and
climate policy outcomes.
One approach to designing climate-related adjustments
to monetary policy frameworks would be to apply
forward-looking “stress tests” to asset or collateral
values using a range of climate scenarios. Climate risk
scenario analysis is still a nascent field, and best practices
on the quantification of climate-related financial risks is
still under development. A more operationally expedient
alternative, depending on data quality, may be to apply
less sophisticated tools such as precautionary haircut
adjustments.
Using external data providers for risk identication
has operational benets. In principle, this is a pragmatic
way for central banks to align with best market practices
and standards. A precondition for reaping this benefit, to
which the central bank can contribute, is the existence of
best practices and market standards that are demonstrably
robust, in terms of methodologies, data coverage, data
quality management, and accuracy. Currently, climate
data are not always available at the issuer or asset level,
and may need to be approximated using economic sector
averages to achieve full coverage. This creates challenges
and trade-offs for central banks that typically necessitate
more granular (asset-by-asset or issuer-by-issuer) data.
An external, harmonised climate-related taxonomy
of activities can help in implementing climate risk
measures in a credible and transparent manner. However,
taxonomies are not without limitation from a monetary
policy operation perspective: they are activity-based, which
may be hard to map with cross-sector firms; they are static
by nature and run the risk of not capturing risk properly,
for instance climate-disruptive technology firms; their
revision process can be a source of difficulties for central
banks that have plugged their categories and thresholds
into their risk control framework, for instance. Alternatively,
a central bank’s categorisation of assets, firms or sectors
could be based on methodologies developed by external
data providers that have identified carbon-intensive sectors
and physical risk criteria, and provide carbon-related data
on the relevant firms. Categorisation could also be based
on a list of national accounting sectors. This approach also
comes with drawbacks, such as the treatment of holding
or financial subsidiaries of industrial groups, or of cross-
sector conglomerates.
6.2. Metrics
Central banks need to have a clear view of the right
climate-related metrics to use in order to adjust their
operational frameworks. This choice depends on the
type of climate-related risks that are targeted (physical risk,
transition risk) and the central bank’s own circumstances
(e.g. the mandate and the materiality of these risks for the
economic and financial system in which the central bank
operates). To operationalise climate-related risk measures,
central banks must decide what type of data to use. These
could range from geographical location, norm-based
information (compliance with global standards, e.g. the
Paris Agreement), climate/physical metrics (CO
2
e tonnes,
2°C alignment), and carbon intensities to financial valuations
of climate-related risks or hybrids (see Annex 2 for a brief
overview of metrics).
At the current juncture, in the absence of reliable
and commonly agreed ways of putting a price tag on
climate-related risks, central banks may consider using
non-nancial climate risk metrics as a pragmatic way of
capturing them and designing potential adjustments to
their operational framework.
Norm-based metrics have practical benets, but may
prove less appropriate from a risk perspective. They are
one way of aligning with market standards or government
policies. Norm-based policies can come in different shapes,
for example by excluding issuers from countries that have
not signed the Paris Agreement or aligning criteria with
government taxonomies. They are a relatively simple option
to implement, and are often used as a first step towards
addressing climate-related risks. Norm-based metrics are
typically based on a broad range of incidences that may
be valuable for broad risk assessments. However, it would
not actually be possible to quantify climate-related risks
and risk exposures.
Metrics based on CO
2
e emissions oer the advantage
of being well-dened and measurable, but they only
provide a rough indicator of climate-related risks,
without delivering a financial quantification of the risks
incurred. The GHG protocol for climate-related information
disclosures is a widely recognised standard. This makes CO
2
e
emissions a suitable option for use as a proxy for an entity’s
transition risk exposure. Yet complementary information
is needed for an accurate assessment of climate-related
NGFS REPORT
29
risks: some firms with high current CO
2
e emissions might
actually incur low transition costs or be rather well-prepared
for transition.
More sophisticated ways of assessing transition risks
might involve forward-looking metrics. These typically
include both quantitative and expert judgement to
evaluate the “transition readiness” of economic agents.
While they can serve to evaluate assets used in central
bank frameworks, these metrics and methodologies are
still evolving and subject to additional uncertainties.
One potential approach is to align an investment portfolio
with target temperature trajectories. Under this approach,
the climate-related risk associated with a given portfolio
is measured by that portfolio’s alignment with a given
temperature target. One advantage of this “alignment”
approach is that it leads to a more direct climate-related
risk measurement without attempting to translate climate
risks into financial risks. A simpler, though still dynamic
alternative to these advanced approaches is to use
CO
2
e emission variations over time.
18
At the current juncture, relying solely on ratings provided
by external credit rating agencies is not sucient to
capture climate-related risks. This would require those
agencies to take climate-related risks into account more
explicitly in their methodologies, which may be difficult
given the shorter time horizon considered for credit risk
compared to transition risk. For central banks, using credit
ratings directly would typically be easy operationally and
probably uncontroversial as credit ratings are often used
already as a key input in their operational framework.
6.3. Data
Central banks may wish to develop policies to
monitor and manage issues surrounding data quality
and availability. Data on climate-related risks, such as
entities’, projects’ and individuals’ emissions as well as their
respective exposure to climate-related risks, are evolving
rapidly. They differ from standard financial data in terms
of quality, consistency, and availability. Virtually all the
possible adjustments that can be made to operational
frameworks to capture climate-related financial risks
are crucially dependent on reliable and consistent risk
identification and assessment, for which high-quality data
are a critical input.
Data availability and accuracy currently act as
constraints for climate-related risk metrics, as data may be
available only partially or only for certain market segments.
Furthermore, diverging data scopes and methodologies
across different providers can lead to a divergence of metrics
or different degrees of reliability. Self-reported metrics on
emissions may also diverge from metrics calculated by
private data providers based on a bottom-up assessment
of the entity’s business model. The use of multiple metrics
and data providers, as well as their selection, is crucial for
ensuring that data are sufficiently trustworthy to be used
for policy purposes. Central banks cannot meet the data
gap alone, and a collective effort is needed to make reliable
climate data more widely available.
19
Against this backdrop, central banks that are able and
willing to adjust their operational frameworks need to
dene their tolerance to data uncertainty. As mentioned
in Chapter 5, they might wish to play a catalyst role in
improving data quality and availability by introducing
data disclosure requirements for entities with which they
interact in monetary policy operations.
For central banks, discrepancies in disclosures between
larger, listed organisations and smaller ones, or across
economic sectors create challenges. For central banks,
using incomplete data may give rise to level playing field
issues, unless they develop, for operational purposes,
statistical approaches (e.g. internal approaches) to overcome
or circumvent data gaps. Fostering initiatives to improve
disclosures can help avoid the unequal treatment of certain
sectors or firm sizes. In doing so, attention needs to be
paid to the fact that disclosure requirements might place a
disproportionally heavy burden on smaller firms. Authorities
with a relevant mandate could grant these smaller firms
a longer period to enhance disclosure requirements or
introduce size thresholds for compulsory disclosures (as
in the EU NFRD). They could also request information in a
way that is proportionate to sectors’ exposure to climate
change and thus to the amount of climate risk that entities
are exposed to.
18 See Ehlers et al. (2020) for a proposal.
19 See, for example, Financial Stability Board (2020).
NGFS REPORT
30
6.4. Balancing trade-os
The high standards for risk measurement accuracy
typically favoured by central banks cannot currently be
met when it comes to climate-related risk assessment.
As accountable public institutions, central banks favour and
design risk frameworks that meet the highest standards. Yet
there is no clear consensus on what the most appropriate
metrics are; data gaps exist, and modelling climate risks is
challenging. Currently, operationally feasible approaches
to adjusting operational frameworks may have some
limitations in terms of their accuracy.
Against this backdrop, central banks face trade-os if
they decide to address climate-related risks. On the one
hand, there are legitimate reasons for them to wait for a
consensus to emerge on the most accurate approaches to
climate-related risks and for all conditions to be in place
for a fully informed risk assessment. Indeed, as public,
accountable institutions, central banks differ from private
financial institutions. It may prove risky from a reputational
and legal perspective to introduce mitigation measures
for climate-related risks without sufficient knowledge.
If they turn out to have been ill-informed, produce
unanticipated side effects, or adversely impact monetary
policy transmission, central banks’ credibility could be at
stake and they may be exposed to increased legal risks.
On the other hand, climate-related risks may currently
lie unchecked on their balance sheets, possibly even
more so for central banks that have implemented
large-scale crisis measures. The fact that climate-related
risks are difficult to measure accurately and are radically
uncertain would actually point towards central banks
adopting gradual, predictable and precautionary risk
protection measures. Given these considerations, as well
as the magnitude of the challenge ahead, central banks
stand to benefit from experimenting with pilot projects on
specific portfolios or asset classes for climate risk mitigation,
as a starting point. They may also benefit from sharing the
ideas and knowledge they gained from adapting their
operational frameworks to climate risks (see Annex 3). This
could inform and pave the way for timely risk management
measures.
NGFS REPORT
31
The range of options a central bank faces when
considering whether to factor climate-related risks
into its operational framework is potentially very large.
Adjustments could, in principle, be considered in all the
main operational policies that central banks implement
for monetary policy purposes: credit operations, collateral
and asset purchases.
For reasons of eciency, only a few stylised options are
reviewed here. These were chosen because they are deemed
sufficiently generic and relevant across different central banks.
Most options leverage existing central bank tools. However,
some would imply departing, more or less significantly, from
their standard objectives or design. They cover the three
main operational policies mentioned above (see Table 3).
Their choice was informed by proposals published in recent
months in various forums (see Annex 4 for a list).
This review is without prejudice to each NGFS central
bank’s individual assessment. It is based on the criteria
defined in Chapter 3 and leverages the practical expertise
and experience of NGFS central banks. These assessments
are neither recommendations nor indicative of members’
preferences. Rather, they provide a first collective attempt to
compare different options in a structured and consistent manner.
Since operational frameworks vary signicantly across
jurisdictions, the assessment remains at a high level.
In practice, a thorough assessment would require detailed
specifications to properly account for each central bank’s
preferences and circumstances.
1.1. Targeted credit operations
Description of the measure
Central banks could adapt their credit operations to target
climate-related risks. Climate-adjusted targeted credit
operations could steer bank lending towards projects
and agents that are less exposed to climate-related risks
or that aim to finance climate change mitigation efforts.
In practical terms, such climate-related adjustments could
comprise making the conditions for central bank operations
dependent on compliance with climate-related credit
lending benchmarks, desired changes to the collateral pools
for targeted operations, and/or specific eligibility criteria
for monetary policy counterparties. Different combinations
of these parameters could be considered, depending on
policy objectives, the structure of financial markets and
other local circumstances.
Adjusting pricing to reect counterparties’
climate-related lending
The central bank could offer attractive interest rate
conditions to encourage commercial banks to increase
their lending to projects and agents that actively contribute
to climate change mitigation or adaption and decarbonise
their business model.
Adjusting pricing to reect the composition
of pledged collateral
The central bank could offer lower (higher) interest rates
to counterparties pledging low-carbon (carbon-intensive)
assets as collateral.
Adjusting counterparties’ eligibility
Central banks could make access to lending facilities
conditional on a counterparty’s exposure to carbon-
intensive, low-carbon or green investments and/or sectors,
or introduce a requirement for counterparties to disclose
their climate profiles.
Table 5. Selected stylised options for adjusting
operational frameworks to climate-related risks
Targeted credit
operations
Adjust pricing to reflect counterparties’
climate-related lending
Adjust pricing to reflect the composition of
pledged collateral
Adjust counterparties’ eligibility
Collateral Adjust haircuts
Negative screening
Positive screening
Align collateral pools with a climate-related
objective
Asset purchases Tilt purchases
Negative screening
Annex 1. Detailed review of options
NGFS REPORT
32
Consequences for monetary policy
eectiveness
Adjustments to credit operations, if they are
specifically targeted at low-carbon and green
investments, may alter the usefulness of such schemes
for monetary policy transmission to a wide set of
economic agents. Indeed, depending on their actual
design, such adjustments could result in a curtailing of
SMEs’ and households’ access to bank lending, as these
agents are less well equipped to disclose and certify that
their investment needs or projects are climate-friendly,
unless central banks and/or financial intermediaries
are mindful to this particular issue. By contrast, larger
corporates, including those with more carbon-intensive
business models, are often already obliged to disclose
climate-related information. Because of these different
disclosure practices across economic agents, climate
adjustments to credit operations could asymmetrically
benefit those banks that service large corporates, as well
as those corporates themselves, potentially distorting
their market share.
More generally, depending on how stringent their design
is, climate-related adjustments may conflict with the policy
objectives of existing central bank facilities, for instance if
they result in a reduction of the maximum amount banks
can borrow. This consideration would be particularly
important in times of crisis, when central bank targeted
lending schemes are typically deployed.
Contributions to mitigating climate change
Credit targeting has the potential to tackle what some
see as the existing carbon bias in both bank lending
and corporate bond markets.
1
It is hoped that banks
would pass better funding conditions on to low-carbon
and greener borrowers, thereby lowering the overall cost
of low-carbon and green investments.
1 See Popov and Haas (2020) and Matikainen et al. (2017).
Figure 5. Targeted credit operations. Comparison of climate-related measures
ADJUSTING PRICING TO BENCHMARKS
ADJUSTING PRICING BASED ON COLLATERAL
ADJUSTING COUNTERPARTIES’ ELIGIBILITY
CONTRIBUTIONS TO MITIGATING
CLIMATE CHANGE
CONSEQUENCES FOR MONETARY POLICY
EFFECTIVENESS
EFFECTIVENESS AS RISK PROTECTION
MEASURES
OPERATIONAL FEASIBILITY
POSITIVE
STRONGLY POSITIVE
MINIMAL
NEGATIVE
STRONGLY NEGATIVE
NGFS REPORT
33
Adjusting credit operations through counterparty
eligibility criteria could inuence lending behaviour,
encouraging lenders to originate or invest more in
low-carbon and greener assets. If carbon-intensive funding
were to weigh on banks’ ability to access central bank
credit operations, this measure may lower demand for
such assets and increase their funding costs. Conditions
based on counterparties’ disclosure or climate strategies
may enhance market practices, promote transparency,
and encourage the inclusion of climate-related risks in
banking operations.
Depending on the structure of nancial markets in
which a central bank operates, combining the different
adjustments could maximise the capacity of credit
operations to steer funding towards low-carbon projects. On
the other hand, greater complexity may discourage uptake
of the programme, in particular by smaller commercial
banks.
Eectiveness as risk protection measures
In principle, adjustments to credit operations may
help to address both physical and transition risks.
The parameters modifying the eligibility of collateral
and/or counterparties could be efficient in addressing
climate-related risks in case they are not properly accounted
for in the market. Central banks can protect their own
balance sheets by increasing their funding price or reducing
the funding available to banks exposed to climate risks and/
or for collateral exposed to climate-related risks.
Operational feasibility
Climate-specific adjustments may increase the
complexity of central bank facilities. Past experience
has shown that the higher the complexity of adjusted
targeted lending schemes, the lower the likely take-up
by banks, and the less effective they may be in achieving
their objectives.
Climate adjustments to central bank facilities may
involve signicant operational, legal, and reputational
challenges, although the existence of various labels for
identifying green and low-carbon assets could help minimise
some of these operational complexities. While significant
efforts have been made to achieve more standardised
market classification practices, existing climate-related
labels allow for significant degrees of discretion on the part
of issuers and banks. Central banks may consider leveraging
their counterparties’ own climate-related initiatives, rather
than relying on disclosures by SMEs and households.
Box 4
Case study. Bangladesh Bank
Bangladesh is committed to pursuing low-carbon green
development without compromising on the imperative
of faster economic growth and social development.
Development strategies that the Government of
Bangladesh laid down in the Perspective Plan (2010-21)
and the Sixth Five Year Plan (2011-15) articulate a clear
commitment to pursuing sustainable growth. The
country’s vulnerability to floods and cyclones, and to the
threat of large coastal areas being inundated by rising sea
levels as a result of global warming makes sustainability a
prime concern. Financing practices can crucially influence
the pace at which environmentally sustainable output
practices are adopted in the real economy. Cognisant of its
responsibility for establishing socially and environmentally
responsible practices in the financial sector, Bangladesh
Bank, the central bank of Bangladesh, has spearheaded
the adoption and promotion of green banking practices
throughout the financial sector, towards safeguarding
environmental sustainability (Millat, 2012).
Bangladesh Bank’s legal mandate is defined as: “[…] to
manage the monetary and credit system of Bangladesh
with a view to stabilising domestic monetary value
and maintaining a competitive external par value of
the Bangladesh Taka towards fostering growth and
development of [the] country’s productive resources in
the best national interest.”(Bangladesh Bank Order, 1972,
Bangladesh Bank (Amendment) Act, 2003). …/…
NGFS REPORT
34
Though sustainability is not officially part of its mandate,
Bangladesh Bank is among the pioneers worldwide in
proactively using its monetary policy toolkit to green
Bangladesh’s financial system. In this vein, it is mobilising
funding to adopt sustainable low-carbon practices in
the private sector which constitutes over four-fifths of
the economy (Kazemi, 2015). While Bangladesh Bank’s
green central banking activities cover almost the entire
range of potential policies (Suttor-Sorel, 2017), this case
study focuses on its green refinancing programmes.
These programmes work by making funding available
to financial institutions and banks at lower rates provided
that lending is directed towards targeted segments of the
economy (Vaze et al., 2019, Barkawi and Monnin, 2015).
Bangladesh Bank launched its first green refinancing line
in 2009, with an initial focus on solar energy, biogas, and
effluent treatment projects, but since then, the scope
of the Refinance Scheme for Environment Friendly
Products/Initiatives has been continuously expanded
(Barkawi and Monnin, 2015). The size of the fund was
doubled recently from USD 25mn to USD 50mn and now
covers 55 product lines belonging to nine categories. The
Refinance Scheme for Environment Friendly Products/
Initiatives was followed by a Refinance Scheme for Islamic
Banks in 2014 to promote lending by these institutions
to green segments of the economy.
More recently, Bangladesh Bank has launched a Green
Transformation Fund of USD 200mn targeted at the export-
oriented textile and leather industries. Established in
2016, the Green Transformation Fund provides funding
for sustainable initiatives, including energy generation
and waste management. Banks accessing the Green
Transformation Fund can borrow from Bangladesh Bank
at LIBOR+1% and are expected to lend to the private
sector with a margin of between 1.00% and 2.00% of
the cost of borrowing (Vaze et al., 2019). The tenor of the
loans is 5 to 10 years, a horizon that is suitable for many
sustainable investments. The rather long tenor of the loans
fills an important gap in Bangladesh’s financial system,
where the corporate bond market is still underdeveloped
and traditional bank lending is usually only provided for
maturities of up to 5 or 7 years (Barkawi and Monnin, 2015).
In 2019, the scope for the USD component of the Green
Transformation Fund was expanded to cover all export-
oriented industries, and in April 2020 an additional fund
of EUR 200mn was introduced alongside the USD 200mn
facility. The euro component of the Green Transformation
Fund has also given scope for importing green capital
goods and can be used to import industrial raw materials
for use in all manufacturing enterprises.
Given the dominance of bank lending in Bangladesh’s
financial system alongside informal funding sources
(Barkawi and Monnin, 2015), Bangladesh Bank’s green
refinancing activities have been, and are likely to remain
successful in promoting funding for sustainable projects.
By providing incentives for commercial banks to spot
new, green loan opportunities, green refinancing lines
are a powerful tool to catalyse the transition towards a
sustainable financial system (Barkawi and Monnin, 2015).
The green initiatives and green practices of Bangladesh
Bank have led to the introduction of a Sustainable Finance
Policy, which has structured not only Green Taxonomy
but also Sustainable Finance Taxonomy. Sustainability
Rating by Bangladesh Bank is now in practice to assess
performance of financial institutions on environmental,
social and governance attributes. Looking ahead,
Bangladesh Bank is currently working on developing
Sustainable Banking Policy and Sustainability Reporting
along with Green Branch Policy, Green Bond Policy
and other related policies aimed at further greening
Bangladesh’s financial system.
NGFS REPORT
35
1.2. Collateral
Four generic adjustments of collateral frameworks to
climate-related risks are examined: (i) adjusting haircuts;
(ii) adapting collateral eligibility criteria, with negative
screening; (iii) adapting collateral eligibility criteria, with
positive screening; and (iv) aligning collateral pools of
counterparties with sustainability objectives. These
measures would produce effects of varying intensity
depending on their calibration.
A) Haircut adjustments
Description
Central banks use haircuts or valuation adjustments to take
the risks of an asset into account. Strictly speaking, haircuts
are typically designed mostly to cover market risks in case
the asset needs to be liquidated. The term “haircut” is also
used in a more general sense to capture other risk-based
adjustments to an asset’s value, including those motivated
by the uncertainty surrounding model-based pricing, where
relevant. Climate-adjusted haircuts potentially come in
different shapes and sizes, depending on their motivation.
Climate haircuts could be used to capture risks that
go unaccounted for in standard haircut calibrations.
For example, higher haircuts could be applied to more
carbon-intensive assets in an attempt to reflect the
additional climate-related risks in future transition scenarios,
especially in cases where markets fail to price them in
adequately. Haircuts on carbon-intensive assets could also
be set above and beyond what is required for risk mitigation
purposes so as to actively discourage the pledging of
carbon-intensive collateral and, hence, indirectly discourage
investment in those economic activities (“penalty haircuts”).
Conversely, a central bank could assign a higher collateral
value (“supporting haircuts”) to assets from less carbon-
intensive sectors. In both cases, such adjustments would
be designed to influence, at the margin, the financing
conditions for assets according to their contribution to
climate-related risks.
Figure 6. Collateral. Comparison of climate-related measures
HAIRCUT ADJUSTMENTS
NEGATIVE SCREENING
POSITIVE SCREENING
2° ALIGNMENT OF COLLATERAL
POOLS
CONTRIBUTIONS TO MITIGATING
CLIMATE CHANGE
CONSEQUENCES FOR MONETARY POLICY
EFFECTIVENESS
EFFECTIVENESS AS RISK PROTECTION
MEASURES
OPERATIONAL FEASIBILITY
STRONGLY POSITIVE
NEGATIVE
STRONGLY NEGATIVE
POSITIVE
MINIMAL
NGFS REPORT
36
Both of these options could be combined by means of a
sliding scale approach.
2
Schematically, for firms operating
in the same economic sector, a haircut add-on (or discount)
could be applied to the assets issued by the comparatively
more (or less) carbon-intensive firm. Such a scheme would
yield a continuous incentive for firms to reduce their
emissions, while safeguarding a level playing field across
sectors and sectoral neutrality.
Moreover, relying on a dynamic approach using metrics
that aim to capture efforts by issuers to address climate-
related risk (e.g. by reducing their carbon emissions) allows
issuers’ commitment to transition to be taken into account.
Adjusting haircuts according to such metrics would factor
in expected changes in issuers’ resilience to transition
scenarios.
appropriately, there is no reason why they should affect
the transmission mechanism for monetary policy beyond
what regular haircuts normally do.
Potential side effects could nevertheless arise, depending
on their calibration. Overly restrictive haircuts may
constrain access to central bank liquidity. Calibrating
climate-adjusted schedules such that the post-haircut
collateral volume remains constant on aggregate can
help circumvent this problem. From this perspective, a
sliding scale of climate-adjusted haircuts may have its
advantages. A haircut discount for low-carbon assets
can result in stronger demand for those assets, fostering
liquidity in those markets and, hence, supporting the
smooth transmission of monetary policy.
Contributions to mitigating climate change
Overall, adjusting haircuts for climate-related risks could
help mitigate climate change at the margin. Experience
has shown that haircuts can influence the preferences of
monetary policy counterparties to invest in, originate,
and pledge some assets as collateral. However, adjusting
haircuts is likely to have a second-order impact relative to
the other measures reviewed, notably negative screening.
Adjusting valuation haircuts to address climate-
related risks will modify the relative value of eligible
assets. It could increase the marginal funding costs of
issuers in the most carbon-intensive sectors and, as a
result, incentivise them to engage more in transition.
It could also induce, at the margin, collateral issuers to
change their debt issuance practices to benefit from
a better haircut. The use of static carbon intensities
can, however, be problematic. This would, for example,
penalise utilities that produce high emissions but have
improved considerably over time and need funding to
make further improvements.
Compared with other haircut options, a sliding scale
approach that penalises and rewards issuers according
to their climate-related riskiness is likely to have the
broadest impact.
2 This suggestion is inspired by fiscal carbon feebates (see, for example, IMF WEO, October 2020). Within an industry, feebates apply, on the one hand,
a sliding scale of fees to firms with above-average emission rates and, on the other hand, a sliding scale of rebates to firms with below-average
emission rates.
Figure 7. Haircut. Example of a climate adjustment
factor using a sliding scale approach
0
10
5
15
20
25
30
35
40
-50-100 0 50 100
150
Sliding scale with three breaks
No sliding scale (constant haircut)
Carbon emissions of firm relative to industry standard (pct.)
Haircut (pct.)
Figure 7 illustrates two haircut regimes. Green line: a xed haircut of 10 pct.
Yellow line: a sliding scale haircut, with a minimum of 4 pct., that increases
slowly until rms’ carbon emissions reach the sectoral standard
and then rises quickly.
Consequences for monetary policy
eectiveness
In normal circumstances, because haircut schedules
have a second-order inuence on the transmission of
monetary policy, climate-adjusted haircuts are expected
to have limited implications for the implementation of
monetary policy. If climate-revised haircuts are calibrated
NGFS REPORT
37
Eectiveness as risk protection measures
Dierentiating haircuts based on climate-related risks
can mitigate residual risks that are potentially not well
captured by existing risk management frameworks.
This is contingent on the existence of robust climate-
related risk measures. Calibrating a haircut in the absence
of commonly agreed methods for measuring the monetary
value of climate-related risks remains a challenge. If
collateral is expected to be liquidated within a short
time horizon, the exposure to climate-related risks may
not materialise.
Applying a haircut discount (add-on) may have adverse
side eects. It can expose central banks to higher (lower)
financial risk if the assets subject to the haircut discount
(add-on) remain, ceteris paribus, exposed to the same level
of financial risk – i.e. if there is no clear correlation between
standard financial risks and climate-related financial risk.
The greater the adjustment, the broader the range of
assets, and the further the adjustments go beyond pure
risk protection calibration, the larger the magnitude of this
potential adverse side effect would be.
From a calibration perspective, the timing mismatch
between the assumption about the time needed to liquidate
the collateral subject to the haircut calibration and the
horizon of climate-related risks makes it challenging to
set haircut levels robust to all scenarios.
Operational feasibility
Haircut calibration requires substantial data inputs, as
central banks typically accept a broad range of assets
as collateral. Existing climate-related data gaps may limit
what central banks are able to achieve in terms of haircut
adjustments.
A well-dened taxonomy of climate-relevant activities or
thresholds can facilitate the design and communication
of climate-adjusted haircuts.
Haircut discounts may be challenging to calibrate
robustly, pending further advances in climate risk
modelling. While there is a growing consensus about
which sectors contribute most to GHG emissions on a
scope 1-3 basis, there is less of a consensus on the exact
risks associated with particular transition pathways or
which sectors are best placed to mitigate climate change
over time. In the absence of this information, the design
of climate-related haircuts may be subject to “pick the
winner” problems. Calibration would also have to take into
account the additional complexity of quantifying climate
risk against other risks (e.g. credit risk), resulting in a single
haircut schedule.
Moreover, haircut calibration may be easier, in operational
terms, for issuer-specific bond collateral than for securitised
bonds or pools of loan collateral.
B) Screening
1.
Negative screening
Description
Negative screening means making some assets ineligible
as collateral, based on e.g. sectoral criteria, compliance with
an external norm and/or climate-related risk characteristics
of issuers or assets.
Negative screening may shield central banks from
certain risk types, but its main rationale goes beyond
risk protection. Indeed, it seems unlikely that an asset
with a positive market value could lose all its value
because of climate change over the typical maturity of
monetary policy credit operations. Negative screening
would typically target assets most exposed to climate-
related risks: this would contribute to mitigating the
central bank’s exposure to climate-related tail risks, such
as transition risks.
There are a range of negative screening strategies.
These can be ratings-based (e.g. ESG ratings), rely on
compliance with a given norm (e.g. countries or companies
not respecting international conventions on climate
change), geographical, or sectoral (e.g. exclusion of fossil
fuels in general). They can be applied at the asset, issuer
or economic sector level.
Consequences for monetary policy
eectiveness
The impact on the smooth implementation of monetary
policy would largely depend on the breadth of negative
screening. If negative screening is broad, applied to a very
NGFS REPORT
38
large range of asset classes or based on excessively stringent
metrics or criteria, it could restrict the participation of
banks in lending operations. That said, screening strategies
are typically most relevant at issuer level and generally
target non-financial corporates. If corporate bonds and
loans generally represent a limited share of the eligible
collateral universe, the overall impact on monetary policy
implementation may be contained.
Regardless of their modalities, screening strategies are
likely to come into direct conict with the level playing
eld and residual risk equivalence principles. Depending
on its design, the measure could impact differently on the
largest and smallest issuers (or debtors) or create distortions
across asset classes.
Contributions to mitigating climate change
Since the eligibility of an asset as collateral typically
influences its liquidity and market price, negative
screening is likely to have a stronger impact than climate-
adjusted haircuts. Screening-based collateral policies
would probably have a greater effect on the market value
of excluded assets than the haircut adjustment approach.
In fact, an exclusion-based policy is equivalent to applying a
100% haircut to the targeted assets. This could support the
transition to a low-carbon economy by raising the marginal
funding costs of excluded firms. That said, depending on its
modalities, negative screening may have adverse impacts
on the transition as they could indirectly disrupt the funding
strategies of issuers engaged in decarbonisation.
Eectiveness as risk protection measures
Negative screening could help limit tail risks, albeit in
a rather crude fashion. It would probably lower the risk
profile of central banks’ balance sheets.
Because it is a rather crude approach, excessively broad
negative screening could adversely aect the central
bank’s protection against nancial risk. The risk efficiency
of negative screening depends on the robustness of the
climate-related risk metric used to discriminate among assets.
Moreover, to the extent that collateral availability may be
curtailed, this approach can drive up concentration risks at
the central bank. Moreover, excluded assets, despite being
those most exposed to climate risk, may well also be those
least exposed to credit risk. While negative screening would
only protect the central bank from assets most exposed to
climate-related risk, it would not address the residual climate
risks associated with collateral that is still eligible. It may
also expose the central bank to reputational and legal risks.
Operational feasibility
Operationally, negative screening is a comparatively
easy way of incorporating climate-related risks into
collateral frameworks. This is especially the case when
screening is based on a taxonomy or norms.
Negative screening would benefit from increased
data availability and a sound taxonomy to identify
carbon-intensive issuers and assets. This is particularly the
case for asset-level screening criteria (vs. norm or economic
sector), as the negative screening of non-compliant issuers
can otherwise lead to a large loss of collateral. The measure
requires a commonly agreed view to identify which activities
or entities are most at risk due to climate change.
2. Positive screening
Description of the measure
Central banks can expand the scope of eligible assets
to include certain types of assets that fund transition
or environment friendly activities (e.g. green bonds or
sustainability linked assets
3
). In practice, this can be achieved
in various ways, e.g. by adjusting the requirements for certain
collateral features (e.g. coupon or principal), increasing the
tolerance for asset complexity, increasing the geographical
coverage of assets, or relaxing mobilisation rules (e.g. pool
concentration limits or own-use of assets). To some extent,
such an approach could leverage market standards (e.g.
for green bonds or green loans, where available)
Consequences for monetary policy
eectiveness
This measure would most likely have little negative effect
on the effectiveness of monetary policy instruments.
3 Sustainability linked assets are marketable assets characterised by a complementary remuneration feature triggered upon achievement of KPIs
linked to environmental objectives by the issuer.
NGFS REPORT
39
If anything, the policy would expand the universe of eligible
collateral without affecting regular collateral and would
most likely be consistent with a smooth implementation
of monetary policy.
Contributions to mitigating climate change
Positive screening helps support the transition of
issuers towards more sustainable business models
and investments. Positive screening would signal
the central bank’s willingness to improve financing
conditions for low-carbon projects through the “eligibility
premium” incorporated in the price of loans issued to
finance these projects. The measure may improve the
liquidity and attractiveness of green and low-carbon
assets, and incentivise monetary policy counterparties
to green their balance sheets. This measure may also
foster the deepening of domestic sustainable bond
and loan markets.
Eectiveness as risk protection measures
Positive screening could result in a higher risk
exposure, to the extent that positive screening implies
higher risk tolerance towards green or low-carbon assets,
except if a better climate profile is strongly correlated
with better credit risk. A priori, the cruder the screening
criteria used by central banks (e.g. sector-level screening
vs asset-level screening), the higher the increased risk
exposure. Moreover, such an approach may face adverse
selection problems.
Operational feasibility
The implementation of positive screening may be
more or less challenging depending on the choice of
criteria used to screen green or low-carbon assets.
Where the market for green or low-carbon assets is less
developed, this measure could face obstacles such as
green asset shortages.
Some variants of positive screening may be easier
to implement, less susceptible to create market
distortions, and less vulnerable to greenwashing than
others. This would likely be the case for measures that
grant low-carbon assets selective derogations from certain
collateral mobilisation rules. For instance, counterparties
may be authorised to pledge “in own use” the assets they
have issued or originated if they are low-carbon or green.
Similarly, where applicable, exemptions to concentration
limits could be granted for low-carbon assets, though this
may have some negative impacts on the overall risk profile.
Yet to eectively implement any variants of positive
screening, the central bank would need to have
clear, transparent, and robust denitions of green
and low-carbon assets, in order to avoid inadvertently
favouring greenwashing.
NGFS REPORT
40
C) Aligning collateral pools with a climate-
related objective
Description of the measure
Counterparties can be required to align the collateral
they pledge so that the collateral pool complies with
a predened climate-related risk metric (e.g. carbon
intensity measure or warming objectives). The metric used
can be static, such as the weighted average carbon intensity
(WACI), or build on forward-looking alignment metrics
4
. These
metrics include: (i) relative gap (in %) between the historical
and projected emissions over a predefined period and a 2°C
emission budget trajectory; (ii) decarbonisation metrics used
as a proxy for portfolio alignment (such as avoided/induced
carbon emissions, or carbon intensity per unit of revenue); and
(iii) synthetic rating based on a range of indicators (economic
activities, investments and R&D expenditures, the entity’s
positioning and strategy regarding the low-carbon transition).
Consequences for monetary policy
eectiveness
Aligning collateral pools may be compatible with the
market-oriented conduct of monetary policy. Indeed,
if counterparties meet the alignment criteria they could
Box 5
1 Bonds issued by the China Development Bank, the Agricultural Development Bank of China, and the Export-Import Bank of China. See FTSE Russell.
2 See PBoC Press release (June 1, 2018).
3 See Fang, H., Y. Wang, and X. Wu (2020).
Case study. The People Bank of China’s treatment of green bonds as collateral
In 2014, the People’s Bank of China (PBoC) launched
the medium-term lending facility (MLF), a 3-to-12-month
credit facility provided against eligible collateral.
This facility accounts for an important part of the credit
the central bank provides to Chinese banks. Before June
2018, the following assets were eligible as collateral under
the MLF: government securities, local government debt,
policy financial bonds,
1
central bank bills, and AAA-rated
corporate bonds.
2
On 1 June 2018, the PBoC announced three changes to
the MLF collateral framework. First, the PBoC expanded
the eligible collateral universe to include green bonds,
bonds issued by SMEs, and bonds issued by agricultural
corporations. Second, the PBoC lowered the credit quality
requirement on all eligible bonds from AAA to AA. Third,
at the early stage of this expansion, SMEs’ bonds, green
bonds as well as SMEs’ and green loans were granted
first-among-equals status.
The measures increased the MLF’s eligible bond collateral
by an estimated 400-600 billion yuan (equivalent to
USD80 billion).
3
This eased the collateral shortage that
small and medium-sized financial institutions were
suffering in the Chinese domestic market.
The changes to the collateral framework qualify as
proactive mitigation of climate change. The expansion
targeted, amongst others, green bonds and loans,
which were previously excluded from the collateral
framework. Furthermore, through its first-among-equals
policy, the PBoC temporarily favoured the funding
of green projects. In this way, the changes to the
collateral framework signalled that, ceteris paribus, the
PBoC would differentiate between bonds with similar
characteristics depending on fund allocation, and, in
so doing, make the financing of the green economy
more attractive.
The changes to the collateral framework cannot be seen
as risk-protective. The reason for this is that the PBoC did
not associate green bonds with a better level of credit
quality than comparable non-green bonds.
4 Methodology and discussion are detailed in TCFD Knowledge Hub, Portfolio Alignment Team’s report (2020). A practical experiment can be found
in Oustry et al. (2020).
NGFS REPORT
41
freely manage and optimise the composition of their
collateral pool. In this context, an option of this kind may
be more consistent with the typical principles underpinning
monetary policy implementation than screening policies.
However, depending on the availability of aligned
assets, counterparties may face challenges in exibly
constituting an aligned collateral pool and managing it
over time. The central bank would need to carefully assess
the amount of eligible collateral that could be considered
aligned under different alignment constraints, before
implementing such a new requirement. Contingent on
the alignment technique and the actual transition efforts
of issuers, the requirement may imply a significant decrease
in corporate assets in the pledged collateral.
Contributions to mitigating climate change
Contingent on the robustness of alignment methods,
requesting counterparties to align collateral pools
would imply that a central bank’s collateral does not
hamper the transition to a low-carbon economy. Besides,
it would incentivise banks and market participants to
gradually adjust their own investment preferences towards
low-carbon assets. This measure would also boost awareness
surrounding the existence of non-aligned assets, which
are more likely to be at risk of becoming stranded. More
generally, this measure could potentially foster a gradual
change in relative prices between assets with different
degrees of carbon intensity, leveraging on market forces.
Compared to more direct measures like negative or
positive screening, aligning collateral pools may,
however, contribute only gradually to reducing climate-
related risks. Indeed, alignment approaches consider
decarbonisation as a process over time, rather than as a
point-in-time achievement
5
.
Eectiveness as risk protection measures
Alignment requirements can reduce exposure to
transition risks. They can reduce the aggregate risk
exposure while safeguarding the risk mitigation benefits of
collateral pool diversification. Yet forward-looking alignment
methodologies are still work in progress and not mainstream
in portfolio management.
Operational feasibility
Unlike other climate-adjusted collateral measures,
pool alignment requirements would typically
constitute a new type of rule and would most likely
increase complexity. Compliance with the criteria
would need to be monitored, adding an operational
burden for both the central bank and monetary policy
counterparties.
Central banks would have to decide on which alignment
criteria to use. Operational feasibility would depend
significantly on the availability of reliable data sources and
the metrics used. The choice of these criteria would have
to strike a balance between the relevance of the metric,
and its coverage of eligible collateral.
1.3. Asset purchases
Compared to collateral measures, asset purchases
create a more direct exposure to climate risks for central
banks, all the more so if assets are held to maturity. Since
climate-related risks are probably not yet fully captured
in standard credit risk metrics, central banks could
consider adopting climate-related adjustments to their
asset purchases in the implementation of their monetary
policy framework.
Central banks considering explicitly taking account
of climate considerations in their asset purchases
can draw insights from experience gained from the
implementation of sustainable and responsible
strategies in their non-policy portfolios.
6
Two stylised approaches could be considered for adjusting
asset purchases to climate-related risks: (i) introducing asset
purchase tilting; and (ii) negative screening.
5 See, for example, Raynaud et al. (2020).
6 Work done in the NGFS on non-monetary policy portfolio strategies shows that: (i) thematic strategies via green bond purchases and best-in-class
strategies applied to corporate issuers have proven useful in contributing to climate change mitigation; (ii) integrating Sustainable and Responsible
Investment (SRI) indicators into the investment decision process tilts capital orientation towards assets that are less carbon-intensive, thus protecting
the portfolio from some climate-related risks.
NGFS REPORT
42
A) Tilting purchases
Description of the measure
Asset purchases could, in principle, be tilted towards
better-performing issuers or assets according to the
climate-related criteria applied, although this would
move away from traditional central bank interpretations
of market neutrality. This can be done by using a climate,
financial (including ratings) or hybrid metric. Tilting is
likely to be more relevant for corporate than for sovereign
assets.
Consequences for monetary policy
eectiveness
Tilting may have less severe consequences for the
available policy space than direct and broad exclusions
resulting from negative screening. If tilting properly
captured climate-related risks and were applied consistently
across all eligible corporate issuers or assets, the measure
could be expected to have only a limited impact on
monetary policy transmission.
Tilting may be less susceptible to legal challenge than
negative screening. This clearly depends on the actual
design of the approach. More generally, if tilting were
implemented in a manner that resulted in a large set of
issuers being systematically excluded from purchases, the
measure could have adverse impacts on market neutrality
and even monetary policy transmission, particularly in
times of market stress. This latter risk might be most
marked in regions where economic activity is relatively
more dependent on the sectors most affected. It would
therefore be important to design tilting measures so that
they take into account individual companies’ performance
and, ideally and to the extent practicable, their future
decarbonisation plans.
Contributions to mitigating climate change
Tilting assets purchases could, in principle, be eective
in contributing to climate-related risk mitigation.
Depending on how it is designed in detail, the measure
could reduce central bank financing of issuers that are
highly carbon-intensive and/or incentivise eligible issuers to
consider and reduce their climate risk footprint, particularly
where corporate asset purchases take place in primary
Figure 8. Asset purchases. Comparison of climate-related measures
TILTING PURCHASES NEGATIVE SCREENING
POSITIVE
STRONGLY POSITIVE
MINIMAL
NEGATIVE
STRONGLY NEGATIVE
CONTRIBUTIONS TO MITIGATING
CLIMATE CHANGE
CONSEQUENCES FOR MONETARY POLICY
EFFECTIVENESS
EFFECTIVENESS AS RISK PROTECTION
MEASURES
OPERATIONAL FEASIBILITY
NGFS REPORT
43
markets. The measure could, in principle, be designed or
calibrated to also reduce climate-related risks for central
banks’ balance sheets, as well as to support a shift in
financing towards low-carbon issuers and assets. Tilting
over time should lead the composition of asset holdings
to gradually shift towards lower-carbon issuers, assets
and sectors.
In the short term, this measure would principally have more
of a signalling effect. It could serve to increase demand for
low-carbon assets, helping to ease access to market financing
for firms that are transforming their business models and
operations to shield them from adverse transition risk.
Eectiveness as risk protection measures
Tilting is a relatively sophisticated way to reduce
climate-related risks. It would likely imply taking climate-
related risks as well as financial risks metrics into account.
The risk efficiency of the measure would largely depend on
the ability of central banks to properly capture an issuer or
asset’s carbon intensity and/or climate-related risk exposure.
Decarbonising asset purchases may also increase risk
protection by reducing a central bank’s exposure to the
assets that are most at risk of being stranded, albeit at some
cost in terms of a loss in standard credit risk diversification.
The further behind issuers are in terms of adapting their
business to climate change, the more stringent the impact
on the purchasable universe could be.
Operational feasibility
Guiding actual purchases using climate-related risk
metrics would likely be a source of additional operational
complexity. Central bank asset purchases typically have to
comply with predefined and largely public rules. Adding
climate-related factors to these rules may have a bearing
on central banks’ scope to engage in agile trading. The
degree of operational complexity would depend on the
metrics and methodologies chosen.
In their purchases according to predened rules, central
banks may employ some level of discretion when they
choose between comparable eligible assets in their
daily transactions. These discretionary elements of
day-to-day investment decisions are typically not public,
for sound market functioning reasons. Asset purchase
investment decisions are typically less transparent at an
individual security level than decisions related to pledged
collateral which rests on the principle of ex ante clarification.
Central banks are unlikely to be in a position to make their
methodologies entirely public.
Central banks should approach tilting in the knowledge
that their action will likely be standard-setting.
This action can positively influence markets through signalling.
In designing tilting methodologies, central banks should
leverage, to the largest extent possible, robust and commonly
agreed metrics and references (taxonomies). This would
prevent abrupt shifts in the risk perception of those issuers
most likely to be affected. More generally, clear communication
and allowing sufficient time for market participants to adjust,
may serve to minimise any concomitant market disruption.
B) Negative screening
Description of the measure
Climate-related risk criteria can be used to screen and
exclude some assets, issuers or sectors from purchases.
Negative screening can be implemented at the inception
of purchases, or applied to new net asset purchases, and/
or when reinvesting maturing bonds. Depending on its
calibration, the effect of screening may vary considerably.
Negative screening can be calibrated so that very few or
no actual exclusions are made initially. In this case, issuers
are incentivised to remain compliant with screening rules
to avoid being excluded in the future.
Consequences for monetary policy
eectiveness
The loss of policy space would depend on the relative
importance of the excluded sectors, issuers or assets in
the economy. Direct exclusion of some sectors or issuers
from asset purchases could reduce the effective monetary
policy space. A priori, the more stringent the screening, the
greater the loss of potential policy space.
By construction, negative screening would not be
compatible with the willingness to homogenously
distribute liquidity across economic sectors. That said,
although the stimulus to the excluded sectors or issuers would
be reduced, it would not be entirely eliminated, as those sectors
would continue to benefit from portfolio rebalancing effects.
NGFS REPORT
44
Contributions to mitigating climate change
Exclusion resulting from negative screening would likely
have more direct and larger impacts than tilting. It would
directly curtail central bank funding of high carbon issuers
or assets. Depending on market circumstances, the loss of
eligibility may also impact negatively on other sources of
financing for these issuers.
That said, depending on the climate-related metrics and
the stringency of the scheme, negative screening applied
to asset purchases faces the same problems as when it is
applied to collateral: it may negatively influence transition
efforts by carbon-intensive issuers, depending on the
screening criteria.
Being prominent and authoritative market participants,
central banks would need to be mindful that negatively
screening some issuers could be seen as a benchmark or
good practice by other participants, while current climate-
related data and risks metrics are not yet fully stabilised
and standardised.
Moreover, a central bank’s decision to use negative
screening in monetary policy asset purchases would
likely come under intense public scrutiny and would need
to be well justified if any legal and reputation challenges
are to be avoided.
Eectiveness as risk protection measures
Negative screening lowers a central bank’s exposure
to climate-related financial tail risks more than tilting.
Nevertheless, the reduction in risk would vary greatly
in magnitude and scope with the screening approach
and the specific climate-related risk criteria on which
it would rest.
That said, negative screening might not be the most
ecient way to reduce climate-related risks. Negative
screening may, in some instances, expose central banks
to trade-offs between investing based on climate-related
criteria and standard credit risk considerations. Moreover,
it would improve the central bank’s risk profile only to
the extent that climate-related risk mitigation offsets
the lower standard risk diversification. However, unlike
in the case of commercial banks, diversification of risks is
not the primary concern in the design of asset purchase
programmes.
Operational feasibility
Screening assets or issuers in asset purchase portfolios
requires a denition of climate-related risk criteria and
thresholds. A broad range of options could be considered,
such as norms, geographical locations, economic sectors
and carbon metrics. To ease its design and implementation,
negative screening could be based on pre-established and
widely used climate standards or norms.
While so-called “high stake” economic activities are rather
well-identified, justifying negative screening would require
an objective approach, given the current lack of harmonised
standards. If screening were conducted at the issuer or
asset level, the need for granular data could be relatively
substantial.
NGFS REPORT
45
Box 6
Case study. The Riksbank’s norms-based negative screening
and enhanced disclosure of carbon data
On 25 November 2020, the Riksbank decided to apply
norms-based negative screening to its purchases of
corporate bonds issued by non-financial companies.
Effective from January 2021, the Riksbank may exclude
bonds from issuers that do not comply with universal global
standards and norms for sustainability. The principles in the
UN Global Compact represent one example of standards
and norms that the Riksbank applies in this context.
The Riksbank must ensure that the conduct of its monetary
policy operations is efficient and that the Riksbank is
economical with public finances. It is therefore expected
to manage the risks that arise from its operations, in this
case the purchase of corporate bonds, including risks to the
Riksbank’s financial position. Sustainability considerations
can form part of the Riksbank’s general risk assessment in
relation to its purchases of corporate bonds. Norms-based
negative screening will form part of the risk assessment on
the premise that companies in breach of universal norms-
based principles are riskier compared with other companies.
This approach also allows the Riksbank to manage public
funds in a way that is in line with the Swedish state’s core
values. Sustainability considerations complement existing
eligibility criteria already applied to the programme, such
as the minimum credit rating criterion.
Sustainability considerations and negative screening will
apply without compromising on the monetary policy aim
adopted for the asset purchase programme. Any decision
on exclusions will be subject to careful monetary policy
and market functioning considerations. The Riksbank
will follow the principle of market neutrality within the
eligible universe that meets sustainability-related risk
criteria.
The Riksbank makes sustainability assessments on its own,
but also use sustainability data produced by an external
supplier as inputs in its own assessments.
In addition to using norms-based negative screening
when purchasing corporate bonds, the Riksbank also
intends to measure and report on the carbon footprint
made by its corporate bond portfolio. This will enable
the Riksbank to promote the reporting of climate factors
in general and create incentives for financial institutions
such as banks and asset managers to disclose the
carbon footprint of their assets and investments. The
increased demand for better disclosure by financial
institutions may in turn create incentives for companies
and other organisations to measure and disclose their
GHG emissions.
NGFS REPORT
46
Currently, a variety of metrics exist to assess climate-related
risks (of a given asset, issuer, portfolio or project) and are
used by investors to calibrate or adjust their investments.
GHG emissions are a key part of all major climate-related
transparency initiatives. Compared to other elements
of relevance to climate change, the measurement and
availability of information on GHG emissions by companies is
relatively advanced. GHG emissions, in quantitative terms, are
usually measured in CO
2
and CO
2
equivalents, respectively.
7
As regards carbon accounting schemes, i.e. the systematic
collection and quantification of GHG emissions, the
standards established by the GHG Protocol market initiative
8
are regarded as the most commonly used framework among
private and public market participants for measuring their
individual carbon footprints. Reporting practices under the
GHG Protocol follow three categories for delineating and
classifying different types of emissions:
• scope 1: direct GHG emissions (occurring from sources
directly owned or controlled by a company/entity);
•
scope 2: indirect GHG emissions from generation of
purchased electricity consumed by the company (i.e. the
emissions physically occur at the power generation facility);
•
scope 3: other indirect GHG emissions (optional
reporting category, allowing for the treatment of all
other indirect emissions which are a consequence of
the activities of the company, but occur from sources
not controlled by the company, e.g. extraction and
production of purchased materials; transportation of
purchased fuels; etc.).
Scope 1 and scope 2 emissions reporting could offer a
very first starting point from which to engage in further
and more in-depth analyses.
While available metrics differ in their methodological
details, they can be grouped into forward-looking and
backward-looking metrics. The metrics used most frequently
in financial markets are backward-looking ones, given
that they are relatively easy to measure, i.e. they capture
a quantitative target.
9
These may include the following:
• absolute greenhouse gas (GHG) emissions;
•
relative GHG emissions (scaled by a relevant factor: sales,
revenues, debt, GDP, population, etc.);
• climate scores (e.g. power generation mix, “sustainable
share”/“carbon-intensive share”).
Backward-looking metrics of this kind may be not ideal in
terms of projecting a future path for gradual alignment with
climate-related targets for a given set of companies or assets,
but unlike forward-looking metrics (e.g. 2°C alignment)
they are easier to measure and compare. Hence, backward-
looking metrics are more operationally feasible from a
central bank’s point of view.
Depending on the climate-related metrics chosen, the
exposure can be derived in absolute or relative terms:
•
carbon footprint calculates the tonnes of GHG emissions
per million euro invested. It expresses the amount of
annual GHG emissions which can be allocated to the
investor per million euro invested in a portfolio;
•
carbon intensity expresses total GHG emissions relative
to the total share of revenue attributed to an investor.
It is expressed in tonnes of GHG emissions per million
euro of revenue. Introducing revenue adjusts for
company size and is therefore a measure of a portfolio’s
carbon efficiency;
•
WACI also calculates the carbon intensity of each
portfolio company in relative terms and scales it based
on its weight in the portfolio. This metric can be used
for comparison with a benchmark, to define reduction
targets and potential ways to decarbonise a portfolio,
as well as for reporting purposes. This metric allows for
portfolio decomposition and attribution analysis, though
it is more sensitive to outliers and favours companies with
7 CO
2
equivalents estimate the global warming potential (GWP) of a given type and amount of GHG, using the functionally equivalent amount or
concentration of carbon dioxide (CO
2
) as the reference; defined in detail in the scientific methodologies used by the UN Intergovernmental Panel
on Climate Change (IPCC).
8 A partnership between the World Resources Institute (WRI) and the World Business Council for Sustainable Development (WBCSD); the GHG Protocol
works with governments, industry associations, NGOs, businesses and other organisations.
9 Forward-looking metrics (e.g. 2°C alignment or physical risk exposure) are more difficult to measure and compare and usually less accessible.
Annex 2. Climate-related metrics
NGFS REPORT
47
higher pricing levels than their peers. The drawback of
this metric is that it provides information on the exposure
or impact of a portfolio at a specific point in time, usually
based on the last reporting period’s emissions and sales
data, without including any forward-looking information.
The key concepts for the most common monitoring
metrics are summarised in the Sustainable and Responsible
Investment Guide published by NGFS in October 2019.
10
While the guide is intended for use in portfolio management,
it is also useful from a more general perspective.
10 See NGFS (2019c).
NGFS REPORT
48
Climate change is a global problem with diverse and
far-reaching implications for monetary policy operations.
Individual central banks are likely to find coordination
beneficial, because sharing ideas and knowledge derived
from each institution’s analyses and assessments can lead to a
better understanding of this type of risk. Additionally, it might
be expected to provide some guidance on how best to adjust
operational procedures to climate risks, if needed, as well as
limit adverse spillovers. And to the extent that central banks
wish to act to mitigate climate change, collective action is likely
to be more effective than “going it alone”, when necessary.
Coordination during the period of transition can be viewed
as particularly beneficial. Central banks are key players not
only in their domestic markets, but often in global markets as
well. Material changes in operational frameworks, especially
among reserve currency issuers, may have substantial
cross-border effects. These potential spillovers to the
global financial system bring attendant risks which must be
carefully considered, and might provide a strong rationale
for coordination, while the central banking community has
a long track record of successful cooperation.
Benets of coordination
The details of the frameworks that central banks use
to meet their operational targets often differ notably
across jurisdictions. But at their core, all frameworks deal
with the central bank’s transactions with the banking
system, both in normal times and in crises, and with the
prudent management of the central bank’s own financial
exposures to its many counterparties. It follows that in
designing climate-related adjustments to their operational
frameworks, central banks will address a similar set of
issues, and so stand to benefit from at least some forms
of cooperation and information sharing.
Adopting common standards for climate-related financial
disclosures would ease the burden on central banks and
financial institutions alike, and aid cooperation, as central
banks rely both on internal and external risk assessments.
Central banks may also find value in reaching broadly
similar views on matters such as the haircuts required to
achieve risk equivalence on climate-exposed asset classes.
Counterparty eligibility frameworks
The criteria determining whether an institution is
eligible as a monetary policy counterparty differ across
jurisdictions. There is nevertheless an opportunity to
cooperate, in particular on the definition of climate-related
disclosure requirements as part of the eligibility process.
Reporting standards such as those proposed by the TCFD
could be used as guidance. Central banks’ coordination
efforts would also benefit monetary policy counterparties,
notably global institutions, as similar standards would
lower the reporting burden.
Necessity of coordination
Asset prices
Changes to any aspect of central banks’ operational
frameworks that tend to favour sustainable assets, or
disfavour carbon-intensive assets, have potential price
effects: the “eligibility premium”, as it is known, for assets
that can be used to obtain central bank money through
Lombard facilities; and “local supply” effects where
investors lack close substitutes for assets the central
bank purchases outright. The direct pecuniary impact
that operations have on other agents may be amplified
in undesirable ways when sustainable assets are already
scarce or markets for them are underdeveloped. Spillovers
could be large if a central bank were to make carbon-
intensive assets ineligible, or impose unattractive terms
on them, during a stress event, potentially amplifying price
moves. Spillovers of a different variety are conceivable in
cases where central bank operations induce new capital
market issuance (as has been observed with certain
corporate bond purchase programmes) by corporations
located in other jurisdictions.
Collateral frameworks
The rules governing access to central bank money through
standing facilities lies at the core of their emergency liquidity
assistance function and the day-to-day implementation
of monetary policy decisions. Their importance for core
central bank functions means that changes in the rules
Annex 3. Coordinating climate-related adjustments to operational
frameworks
NGFS REPORT
49
must be clearly and carefully communicated to banks and
market participants. A premature narrowing of collateral
frameworks, or poorly communicated changes, could
present stability risks that spill across borders or jeopardise
price stability objectives.
Whether they choose to enact changes in the near term
or plan to wait, central banks will need to monitor how
their counterparties respond to evolving rules. Where
divergent treatment of climate-sensitive assets exists
between jurisdictions, it is possible that collateral
will be repositioned. Central banks that have moved
ahead may find themselves with unexpectedly larger
exposures to sustainable assets that are illiquid and
costly to manage. Those who move more slowly may
find that their exposures are tilted towards assets that
are more at risk of stranding.
Foreign exchange operations
Central banks conduct several types of foreign exchange
operations, such as swaps and repos, besides outright
purchases and sales of foreign exchange. They carry out
these operations for various purposes, such as foreign
currency liquidity management, reserve management,
international financial cooperation, and foreign exchange
interventions. A few central banks also use the exchange
rate as a tool to either directly or indirectly support their
overall policy goals.
The central banks in this group hold a portfolio of safe
and liquid foreign exchange instruments to satisfy policy
objectives. While some central bank foreign asset portfolios
have clear short to mid-term policy goals requiring stability
and low risk levels, others may have more leeway to adopt a
broader scope for sustainability. The need for central banks
to coordinate and adopt common identification standards
and definitions of green bonds would facilitate consistency
with investment goals, risk tolerance and liquidity.
Limiting spillovers
Much of the potential for the adverse spillovers described
above to manifest themselves can be mitigated through
effective coordination that ensures that climate adaptation is
well sequenced, clearly communicated, and which minimises
the scope for counterparties to arbitrage different rules.
Takeaways
In summary, central banks can benefit from maintaining
engagement with developing ideas in the operations space,
including through participation with the NGFS. Closer
cooperation is likely to be valuable amongst regional groups,
and among central banks with similar operational frameworks
(for example, those where standing facilities are routinely used
in policy implementation). However, central banks should
remain at liberty to move at a pace suited to their jurisdiction,
while remaining mindful of the spillovers of their decisions,
and learning lessons from the experience of early adopters.
NGFS REPORT
50
Aglietta M., E. Espagne, and B. Perrissin Fabert (2015)
A proposal to finance low-carbon investment in Europe.
French Ecology, Sustainable Development and Energy
Ministry. Department of the Commissioner general for
sustainable development, No 121, English version.
Alessi, L., E. Ossola, and R. Panzica (2019)
The greenium matters: Evidence on the pricing of climate
risk. Joint Research Centre of the European Commission.
Working Papers in Economics and Finance, 2019/12.
Anderson V (2015)
Green money: reclaiming quantitative easing. Money
creation for the common good. Green/EFA group, European
Parliament.
Arezki and Obstfeld (2015)
The price of oil and the price of carbon. International Monetary
Fund Blog, December 2.
Arndt, C., C. Loewald, and K. Makrelov (2020)
Climate change and its implications for central banks in
emerging and developing economies. South African Reserve
Bank Working Paper Series WP/20/04.
Bangladesh Bank (2020)
Quarterly Review Report on Green Banking Activities of
Banks & Financial Institutions and Green Refinance Activities
of Bangladesh Bank.
Batten, S., R. Sowerbutts, and M. Tanaka (2016)
Let’s talk about the weather: the impact of climate change
on central banks. Bank of England. Working Paper, No 603.
Battiston S. and I. Monasterolo (2019)
How could the ECB’s monetary policy support the sustainable
nance transition? FINEXUS: Center for Financial Networks
and Sustainability, University of Zurich.
Barkawi A. and P. Monnin (2015)
Monetary policy and sustainability. The case of Bangladesh.
Council of Economic Policies, United Nations Environment
Programme. Inquiry working paper 15/02.
Barkawi A. and P. Monnin (2015)
Greening China’s Financial System. Chapter 7: Monetary
Policy and Green Finance: Exploring the Links. Council of
Economic Policies, United Nations Environment Programme.
Country report.
Böser, F. and C. Colesanti Senni (2020)
Emission-based Interest Rates and the Transition to a
Low-carbon Economy. CER-ETH Economics working paper
series 20/337, CER-ETH – Center of Economic Research
(CER-ETH) at ETH Zurich.
Bolton, P. and M. Kacperczyk (2020)
Carbon Premium Around the World., Centre for Economic
Policy Research, London. CEPR Discussion Paper Series, No
DP14567, April.
Bui, B., O. Moses, and M. N. Houqe (2019)
Carbon disclosure, emission intensity and cost of equity
capital: multi-country evidence. Journal of Accounting and
Finance, Vol. 60(1), 47-71.
Brunnermeier M. and J. P. Landau (2020)
Central banks and climate change. VoxEU, CEPR.
Campiglio, E. (2016)
Beyond carbon pricing: The role of banking and monetary
policy in financing the transition to a low-carbon economy.
Ecological Economics, 121, 220–230.
Campiglio, E., Y. Dafermos, P. Monnin, J. Ryan-Collins,
G.Schotten, and M. Tanaka (2018)
Climate change challenges for central banks and financial
regulators. Nature Climate Change 8(6), 462-468.
Dafermos Y., M. Nikolaidi, and G. Galanis (2018)
Climate Change, Financial Stability and Monetary Policy.
Ecological Economics, Vol. 152, 219-234.
Dafermos, Y., M. Nikolaidi, and G. Galanis (2018)
Can Green Quantitative Easing (QE) Reduce Global Warming?
Policy Brief, Greenwich Political Economy Research Centre.
Annex 4. Bibliography and overview of recent proposals
NGFS REPORT
51
Dikau S. and J. Ryan-Collins (2017)
Central banks, climate change and the transition to a
low-carbon economy. New Economics Foundation.
Dikau S. and U. Volz (2018)
Central Banking, Climate Change and Green Finance. Asian
Development Bank Institute. Working paper series, No 867.
Dikau S., N. Robins, and M. Täger
Building a sustainable nancial system: the state of practice
and future priorities. Financial Stability Review, Issue 37.
Banco de España.
Dikau, S., N. Robins, and U. Volz (2020)
A Toolbox for Sustainable Crisis Response Measures for Central
Banks and Supervisors. Grantham Research Institute on
Climate Change and the Environment, London School
of Economics and Political Science and SOAS Centre for
Sustainable Finance, SOAS University of London.
Downar, B., J. Ernstberger, S. Reichelstein,
S.Schwenen, and A. Zaklan (2020)
The Impact of Carbon Disclosure Mandates on Emissions and
Financial Operating Performance. Deutsches Institut für
Wirtschaftsforschung. Discussion Paper, No 1875.
D’Orazio P. and L. Popoyan (2018)
Fostering Green Investments and Tackling Climate-Related
Financial Risks: Which Role for Macroprudential Policies?
Ruhr Universität Bochum. Economics Paper No 778.
D’Orazio P. and L. Popoyan (2020)
Taking up the climate change challenge: a new perspective
on central banking. LEM Papers Series 2020/19, Laboratory
of Economics and Management (LEM), Sant’Anna School
of Advanced Studies, Pisa, Italy.
D’Souza, R. and T. Rana (2020)
The Role of Monetary Policy in Climate Change Mitigation.
ORF Issue Brief No 350.
Ehlers, T., Mojon, B. and Packer, F (2020)
Green bonds and carbon emissions: exploring the case for
a rating system at the rm level. Bank for International
Settlements. BIS Quarterly Review, September.
Fang, H., Y. Wang, and X. Wu (2020)
The Collateral Channel of Monetary Policy: Evidence from
China. NBER Working Paper, No 26792.
Farid et al. (2016)
After Paris: Fiscal, Macroeconomic, and Financial Implications
of Climate Change. International Monetary Fund Staff
Discussion Note, January.
Fender, I, M. McMorrow, V. Sahakyan,
and O. Zulaica (2020)
Reserve management and sustainability: the case for green
bonds? BIS Working Papers, No 849, March.
Financial Stability Board (2020)
The Implications of Climate Change for Financial Stability,
November.
Fisher, P. and A. Kern (2019)
Climate change: the role for central banks Working paper
No 2019/6, King’s College Business School, DAFM Research
Centre.
Flaherty E. (2020)
Green Central Banking: Options for the ECB on Climate
Change. Future of the EU27.
Friede, G., T. Busch, and A. Bassen (2015)
ESG and financial performance: aggregated evidence from
more than 2000 empirical studies. Journal of Sustainable
Finance & Investment, Vol. 5(4).
Haas R. and A. Popov (2019)
Finance and Carbon Emissions. European Central Bank.
Working Paper Series, No 2318.
Honohan, P. (2019)
Should Monetary Policy take Inequality and Climate Change
into Account? Peterson Institute for International Economics,
Working Paper 19-18.
Hsu, P.-H., K. Li, and C.-Y. Tsou (2018)
The Pollution Premium, November.
Intergovernmental Panel on Climate Change (2018)
Global warming of 1.5°C.
NGFS REPORT
52
International Monetary Fund (2019)
Global Financial Stability Report: Lower for Longer, Chapter6–
Sustainable Finance: Looking Farther, October.
Jourdan S. and W. Kalinowski (2019)
Aligning Monetary Policy with the EU’s Climate Targets,
Veblen Institute for Economic Reforms.
Kazemi, M. (2015)
Reorienting Financial Intermediation towards sustainable
nancing: Bangladesh Bank’s approach.
Krogstrup, K. and W. Oman (2019)
Macroeconomic and Financial Policies for Climate Change
Mitigation: A Review of the Literature. International Monetary
Fund. Working Papers, 2019/185.
Lagarde C. and V. Gaspar (2019)
Getting Real on Meeting Paris Climate Change Commitments.
International Monetary Fund Blog, May 3.
Macquarie R. (2019)
A green Bank of England – central banking for a low-carbon
economy. Positive Money.
McConnell, A., B. Yanovski, and K. Lessmann (2020)
Central Bank Collateral as an Instrument for Climate
Mitigation. Potsdam Institute for Climate Impact Research.
McGlade C. and P. Ekins (2015)
The geographical distribution of fossil fuels unused when
limiting global warming to 2 °C. Nature. Vol. 517(7533),
187-90, January.
Martinez-Diaz L. and G. Christianson (2020)
Quantitative easing for economic recovery must consider
climate change. WRI.
Matikainen, S., E. Campiglio, and D. Zenghelis (2017)
The climate impact of quantitative easing. Policy paper.
Grantham Research Institute on Climate Change and the
Environment. Centre for Climate Change, Economics and
Policy.
Millat, K., Rubayat, C. and E. Singha (2012)
Green Banking in Bangladesh Fostering Environmentally
Sustainable Inclusive Growth Process.
Monasterolo I. and M. Raberto (2017)
Is There a Role for Central Banks in the Low-Carbon Transition?
A Stock-Flow Consistent Modelling Approach,
available at SSRN: https://ssrn.com/abstract=3075247 or
http://dx.doi.org/10.2139/ssrn.3075247
Monnin P. (2018)
Central Banks and the Transition to a Low-Carbon Economy.
Council on Economic Policies. Discussion note 2018/1.
Monnin, P. (2018)
Central banks should reflect climate risks in monetary policy
operations. SUERF Policy Note, Issue 41.
Monnin P. (2020)
Shifting gears: Integrating climate risks in monetary policy
operations. Council on Economic Policies, Policy Brief.
Network for Greening the Financial System (2019a)
A call for action – Climate change as a source of nancial risk.
Network for Greening the Financial System (2019b)
Macroeconomic and nancial stability – Implications of
climate change.
Network for Greening the Financial System (2019c)
A sustainable and responsible investment guide for central
banks’ portfolio management.
Network for Greening the Financial System (2020a)
Monetary policy and climate change – initial takeaways.
Network for Greening the Financial System (2020b)
Survey on monetary policy operations and climate change:
key lessons for further analyses.
Oman, W. (2020)
A Role for Financial and Monetary Policies in Climate Change
Mitigation. IMF Blog.
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 Series No 790.
NGFS REPORT
53
PBoC (2018)
PBC Official Answered Press Questions on Expanding Range
of Acceptable Collaterals of Medium-Term Lending Facility
(MLF). Press release of June 1, 2018.
Raynaud, J., Voisin, S., Tankov, P., Hilke, A.
and A. Pauthier (2020)
The Alignment Cookbook: A Technical Review of
Methodologies Assessing a Portfolio’s Alignment with
Low-Carbon Trajectories or Temperature Goal. Institut
Louis Bachelier.
Schoenmaker, D. (2019)
Greening Monetary Policy. Bruegel, Working Paper, Issue 2.
Sen, S. and M. T. von Schickfus (2020)
Climate policy, stranded assets, and investors’
expectations. Journal of Environmental Economics and
Management. Vol. 100 (102277).
Solana J. (2018)
The Power of the Eurosystem to Promote Environmental
Protection. University of Oslo Faculty of Law Legal Studies;
Research Paper Series, No 2018-23.
Suttor-Sorel, L. (2017)
Seven central banks leading on climate change. Positive
Money.
Task-Force on Climate-related Financial
Disclosures (2017)
Recommendations of the Task Force on Climate-related
Financial Disclosures, June.
Task-force on Climate-related Financial Disclosure
Knowledge Hub (2020)
Measuring Portfolio Alignment: Assessing the position of
companies and portfolios on the path to Net Zero.
Tooze A. (2019)
Why central banks need to step up on global warming. Foreign
Policy.
Tucker, P. (2018)
Unelected Power - The Quest for Legitimacy in Central
Banking and the Regulatory State, 556-564.
United Nations Principles for Responsible
Investment (2019)
What is the Inevitable Policy Response?, December.
Van Lerven, F. (2017)
Green central banking in emerging market and developing
country economies. New Economics Foundation.
Vaze P. A. Meng, and D. Giuliani (2019)
Greening the nancial system. Tilting the playing eld. The
role of central banks. Climate Bonds Initiative.
Volz U. (2017)
On the role of central banks in enhancing green finance.
UN Environment, inquiry working paper, 17/01.
NGFS REPORT
54
The technical document Adapting central bank operations to a hotter world: Reviewing some options is a collaborative
effort of the members of the “Scaling up Green Finance” workstream of the NGFS. This document was prepared under the
auspices of the chair of the workstream, Dr Sabine Mauderer (Deutsche Bundesbank). It was steered by Pierre-François
Weber (Banque de France), and Alessandro Calza (European Central Bank), with support from the NGFS Secretariat at the
Banque de France (Amandine Afota and Lisa Biermann) and the chair’s team at the Deutsche Bundesbank.
The chair of the workstream is grateful for the contributions provided by: Béatrice Amaladasse (Banque de France),
Franck Auberger (Banque de France), Jana Aubrechtova (European Central Bank), Pauline Bacos (Banque de France),
Ryan Barrett (Bank of England), Lena Mareen Boneva (European Central Bank), Andreas Breitenfellner (Oesterreichische
Nationalbank), Stéphane Dees (Banque de France), Arturo Diez Caballero (European Central Bank), Claudia Duarte
(European Central Bank), Eleanor Eden (Bank of England), Christopher Erceg (International Monetary Fund),
Bünyamin Erkan (Banque de France), Gianluigi Ferrucci (European Central Bank), Clara Gonzalez (Banco de España),
Vincent Grossmann-Wirth (Banque de France), Daniel Gybas (European Central Bank), Lorenzo Isgro (European
Central Bank), Liliana Jerónimo (Banco de Portugal), Reimo Juks (Sveriges Riksbank), Aliki Kartapani (Bank of
Greece), Mélissa Kasongo Kashama (Banque Nationale de Belgique), Philipp Kuss (Deutsche Bundesbank), Nadia
Laut (Banque de France), Marie Norum Lerbak (Norges Bank), Pauline Lez (Banque de France), Sophie Mages
(Banque de France), Yuji Maruo (Bank of Japan), Thomas McLaren (Bank of England), Roland Meeks (International
Monetary Fund), Katri Mikkonen (European Central Bank), Kathrin Möhlmann (Deutsche Bundesbank),
Marcus Mølbak Ingholt (Danmarks Nationalbank), Francesco Paolo Mongelli (European Central Bank), Yvo Mudde
(de Nederlansche Bank), Mizuki Nakajo (Bank of Japan), Asad Qureshi (International Monetary Fund), Christina Rivellini
(Banque centrale du Luxembourg), Franziska Schobert (Deutsche Bundesbank), Vassilis Spiliotopoulos (Bank of Greece),
Ida Stuhr Sjøblom (Norges Bank), Elod Takats (Bank for International Settlements), Philip Temme (Bank of England),
Aki Tomoda (Bank of Japan), Matthew Trott (Bank of England), Naelle Verniest (Banque de France), Sebastian Weber
(European Central Bank), Mark A. Weth (Deutsche Bundesbank), Christopher Worthington (Bank of England).
The chair would also like to thank the following institutions for their contribution and comments: Bank of Indonesia,
Bangladesh Bank, People’s Bank of China, Reserve Bank of Australia, Reserve Bank of New Zealand, the US Federal Reserve.
Acknowledgements
NGFS
Secretariat
