The Blockchain

Revolution: Decoding
Digital Currencies
After completing this reading, you should be able to:

● Explain how a blockchain-based cryptocurrency system works and compare

cryptocurrencies to conventional money and payment systems.
● Describe elements of a decentralized finance structure, including smart
contracts, tokenized assets, decentralized autonomous organizations, and
decentralized exchanges.
● Define stablecoins and assess their advantages and disadvantages, including
their potential contribution to systemic risk and regulatory considerations.
● Explain the advantages, disadvantages, and potential applications of a central
bank digital currency.

How a Blockchain-based Cryptocurrency System Works

Money, Digital Money, and Payments
Money is a form of social credit because it allows people to exchange goods and
services within a community. In smaller communities, people’s actions can easily be
tracked and recorded, allowing for a communal ledger of individual consumption and
production decisions. In larger communities, where people mostly don’t know one
another, this system doesn’t work, so money is used as a medium of exchange. Money
can be in physical form, like cash, or digital form, like electronic bank transfers. Physical
cash transactions are direct between parties, while digital payments require a trusted
intermediary to process the transfer.

Cryptocurrencies, Blockchain, and the Double-spend Problem of Digital

Money
Cryptocurrencies are digital information transfer mechanisms that can be used as a
form of money and payment system. Unlike conventional money systems,
cryptocurrencies are not controlled by a single entity, but by a group of volunteers called
miners. Miners are responsible for keeping track of all cryptocurrency transactions.
Miners keep their records in a digital ledger known as the blockchain. The protocols that
govern the blockchain are embedded in computer code, and cryptocurrency users must
trust that the rules for using them are fair and won’t be changed without a good reason.
However, managing this digital ledger is not easy. There is a risk that someone could
duplicate the digital money and spend it multiple times, causing what is known as the
“double-spend problem” in the monetary system.

Creating multiple copies of a digital file is easy, allowing the same digital file to be spent
twice. This makes the double-spend problem a challenge in digital money systems.
Each society member would have their own personal money printing press, which is
unlikely to work successfully if they could generate personal copies of digital money
files. Despite not being immune to the double-spend problem, physical currency can be
designed to make counterfeiting sufficiently expensive. The traditional approach to
solving the double-spend problem with digital money is to provide a reliable third party,
such as a bank, to aid in mediating the movement of value between accounts in a
ledger. Bitcoin was the first money and payment system to solve the double-spend
problem for digital money without the help of a trusted third party, and below is an
explanation of how they managed to do that:

The Digital Village: Communal Record-keeping

Cryptocurrencies transaction record keeping can be likened to small communities
whose members know one another. Each member in these communities has a history
of behavior known to others. This history is a virtual database that is shared among
members. The maintenance of this database is a shared responsibility, not solely the
responsibility of one person. The community records members’ contributions here. Each
member’s record is a reputation history based on what individuals have contributed to
the community. In this sense, the credit they receive from the community can be
considered a form of money. Even though it is possible for individuals to fabricate their
history for personal gain, open and shared ledgers are difficult to alter without
communal consensus. This is the fundamental concept behind decentralized finance
(DeFi).

Governance via Computer Code

All social interaction is subject to rules that govern behavior. Behavior in small
communities is governed largely by unwritten rules or social norms. In larger
communities, rules often take the form of explicit laws and regulations. In financial
markets, the rules governing behavior and transactions are important for maintaining
the stability and integrity of the system. In the case of the United States, the Federal
Reserve plays a central role in this system. The Federal Reserve Act of 1913, as well as
other laws and regulations, govern all the actions the Federal Reserve takes. These
rules can change over time as political support for certain changes emerges. Similarly,
in the case of cryptocurrencies, the rules and protocols that govern the monetary policy
and payment processing are built into the code and are difficult to change. Some
individuals see this as an advantage since it provides a clear and unchanging
framework for the system.

How Blockchain Technology Works

While data itself is the centerpiece of operations for any database management system,
cryptocurrencies use a database known as the blockchain. A blockchain is like a ledger
of money accounts with unique addresses. These money accounts work like post boxes
such that anyone visiting the post offices is permitted to see the money balance in each
account, but you need the correct password to access the money. These passwords are
generated automatically upon account opening and are only known to the account user.
The names of these accounts are pseudonymous. Cryptocurrencies are considered
“digital bearer instruments” because possession of the private password determines
their ownership and control. Note that cryptocurrencies are often referred to as “digital
cash” thanks to their similarity to physical cash.

Cryptocurrency transactions are similar to using physical cash since neither permission
nor personal information is required. By downloading a cryptocurrency wallet, anyone
with internet access can communicate with the system’s miners and generate a public
address and private key for their account. This will create a similar front-end experience
to online banking for managing money balances and initiating payment requests.
However, if the private key is lost or stolen, there is no way to recover the funds.

While cryptocurrencies have gained popularity because of their provocative and

glamorous appearance, the real innovation lies in the way in which their databases
operate. Each money account is managed in accordance with a set of computer code
rules that regulate access to the database. Additionally, protocols govern how account
managers are rewarded for updating the database. Proof-of-Work (PoW) and
Proof-of-State (PoS) are two common protocols used in this process. It is worth noting
that some form of gatekeeping is required to prevent the addition of unwanted data to
the database. The primary economic question is whether these protocols can process
payments and manage money accounts more efficiently, cheaply, and securely than
existing centralized finance systems.

Native Token
To record money balances, cryptocurrency uses a monetary unit known as native
tokens. These tokens are similar to foreign currencies. However, computer algorithms,
not a country’s policymakers, determine their value. The potential for capital gains from
appreciating the value of the native token relative to the U.S. dollar often drives the
excitement around cryptocurrencies. However, the fundamentals of a cryptocurrency
that would generate continued capital gains for investors beyond the initial adoption
phase are unknown. Furthermore, while the supply of a specific cryptocurrency, such as
Bitcoin, may be limited, the supply of close substitutes may be unlimited. The total
market capitalization of cryptocurrencies will likely continue to grow. Still, this growth
may be attributed more to the creation of new cryptocurrencies than to a rise in the price
per unit of any given cryptocurrency, such as Bitcoin.

Cryptocurrency Application
Cryptocurrencies, like Bitcoin, have primarily been used as a store of value rather than
a medium of exchange due to their volatility. However, their potential use as a vehicle
currency for international remittances and their decentralized nature, which operates
independently of any government or concentration of power, makes them attractive to
some individuals and organizations. Furthermore, anyone can access Bitcoin payments
freely and without permission, provided they have an internet connection. However, it
also makes it easier for illegal activities and money laundering to occur. El Salvador’s
recent adoption of Bitcoin as a legal tender has aroused interest as a case study of how
a cryptocurrency can perform as a domestic payment system.

Using blockchain technology, Bitcoin operates as a decentralized autonomous

organization (DAO) utilizing open-source computer code on a large network of
computers. Moreover, it has no physical location, is not required to be registered as a
business, and lacks a CEO and traditional employees. Since it has a fixed supply of 21
million BTC and allows anyone to participate, it’s an ideal platform for fast, low-cost,
high-value transactions.

Policy Considerations of Cryptocurrency

Central banks view cryptocurrencies the same way they do (view) foreign currencies,
which governments may consider a threat. Controlling cryptocurrencies would be
difficult due to their nature of being easily accessible and not requiring permission.
Cryptocurrencies may also constrain domestic monetary and fiscal policies by limiting
the amount of seigniorage available to fund government expenses. Furthermore, while
issuing debt denominated in foreign currency, such as the U.S. dollar, may be less
expensive, it may cause problems if the domestic currency depreciates, making it
difficult for debtors to repay and potentially causing a financial crisis. If debt instruments
are denominated in cryptocurrency, there is no way to negotiate with the
cryptocurrency’s decentralized organization (DAO). As a result, domestic regulators
may be more strict in regulating the issuance of cryptocurrency-denominated debt if it
poses a significant risk to the system.

Elements of a Decentralized Finance Structure

Decentralized finance, or DeFi, refers to financial activities conducted on blockchains.
Unlike traditional finance, which relies on intermediaries and centralized institutions,
DeFi uses smart contracts to enable transactions between parties without the need for
intermediaries, which can significantly reduce costs and give parties more control over
the terms of agreements. However, intermediaries continue to play an important role in
areas such as verification and enforcement, and they may not disappear entirely. This
section describes the concepts of DeFi and its implications.
Smart Contracts
A smart contract is a computer program that performs a set of predefined actions
agreed upon by the parties. Nick Szabo first introduced it in the 1990s as a crude
example of a vending machine. Smart contracts enable secure financial transactions
without the use of third parties and can be used for purposes other than traditional
financial transactions. Ethereum is a blockchain platform that supports smart contracts,
which in this case, are a type of account with their own balance that can interact with
the network. Like cryptocurrencies, smart contracts allow for secure and transparent
transactions between untrusted parties without intermediaries. Consider a collateralized
loan as an example. In traditional finance, such a transaction would involve many
parties, including the lender, the borrower, a broker, financial intermediaries, appraisers,
loan servicers, asset custodians, and others.

On the other hand, a smart contract is a computer program that details an agreement
and stores it on the blockchain. It contains information about the loan and the actions
that will be taken based on compliance. The blockchain ensures the contract’s
successful execution, removing the need for any additional parties other than the
borrower and lender.

Asset Tokenization
Asset tokenization involves the conversion of physical assets into digital tokens that can
be used on a blockchain. This allows assets such as real estate to be used as collateral
in smart contracts. However, enforcing blockchain property rights for assets that exist in
physical form remains a challenge for DeFi. Tokens also have non-financial uses, such
as granting voting rights to organizations and creating unique digital artistic images
through nonfungible tokens (NFTs), which serve as a certificate of authenticity. The key
advantage of NFTs is that they use the security of the blockchain to ensure authenticity
and security, rather than relying on signatures that can be forged.

Decentralized Autonomous Organizations

Smart contracts have the potential to fundamentally alter the way institutions are
organized and managed. This applies to investment funds, corporations, as well as
public goods and services. One such application is the decentralized autonomous
organization (DAO), which is an organization represented by computer code and
governed by smart contracts on a blockchain. A popular example is MakerDAO, which
issues the stablecoin Dai and is governed by its stakeholders, who use tokens to make
decisions about protocol changes. Governance refers to the rules that balance the
interests of different stakeholders in an institution. In traditional corporations, the board
of directors plays a critical role in corporate governance, addressing issues such as
agency problems where managers do not act in the best interest of shareholders. DAOs
offer an alternative governance model by encoding rules in a smart contract, replacing
the traditional top-down structure with a decentralized consensus-based model.
Examples include Uniswap, a decentralized exchange, and Aave, a borrowing and
lending platform. These started out with development teams in charge of operations and
decisions. Eventually, they distributed governance to the wider community through the
issuance of tokens, allowing holders to submit proposals and vote on them.

Centralized and Decentralized Exchanges

Nowadays, crypto-assets are mostly traded through a centralized exchange (CEX),
whose operation is similar to a traditional bank or a broker. A client opens an account by
providing personal information and then depositing funds. The client can then trade
crypto assets at prices listed in the exchange. The client’s assets are in the custody of
the exchange; hence the client does not own these assets. As such, all client
transactions are recorded on the database of the exchange rather than on a blockchain.

On the other hand, decentralized exchanges (DEX) use smart contracts for peer-to-peer
trading without intermediaries, allowing traders to keep custody of their funds and
interact directly with smart contracts on a blockchain. Order books, which list buy and
sell orders for a given asset as well as their offer and bid prices, can be used to
implement DEX. CEXs operate like DEXs. However, with DEXs, smart contracts handle
the list and transactions. When all transactions are handled on a blockchain, the order
book is referred to as “on-chain.” Otherwise, it is referred to as “off-chain.” One
particular issue of concern with order books is their potential to slow execution and lack
of liquidity. DEX aggregators and automated market makers (AMM) can solve the
liquidity problem, making trades faster and cheaper. AMMs are currently the most
popular form of DEXs since they offer the best solution to liquidity problems.
Stablecoins
As discussed previously, cryptocurrencies are highly unsuitable as payment instruments
due to their extreme exchange rate volatility. A stablecoin is a cryptocurrency whose
value is linked to an external asset, such as the U.S. dollar, to make it more useful as a
payment method. To accomplish this, the stablecoin must persuade its owners that it
can be exchanged for U.S. dollars at par or at a fixed exchange rate. The aim of this
design is to increase the use of stablecoin by making it more appealing as a payment
option. One significant advantage of using stablecoin instead of a traditional bank
account is that it allows for lower-cost USD payments due to its use of blockchain
technology for more efficient account management and payment processing services,
which can be passed on to customers in the form of lower fees. It can also be viewed as
a method of avoiding certain costs by transferring the transaction outside the regulated
environment.

Financial Stability Concerns

Stablecoins based on the U.S. dollar are similar to money market funds in that the price
of their liabilities is tied to the U.S. dollar. They are also closely related to banks that do
not provide deposit insurance. As pointed out in the 2007-09 financial crisis, even
money market funds are vulnerable to runs if their assets are of poor quality. Similarly,
unless fully backed by U.S. dollar reserves or bills, U.S. dollar-based stablecoins may
experience a bank run. This can happen if a stablecoin is unable to sell its assets at
reasonable prices or raise the funds required to meet its redemption promises. This
poses a risk not only to stablecoin holders but also to the financial system as a whole.

Regulators are also concerned about the possibility of systemic risk if a stablecoin run
leads to a fire sale of commercial paper, as well as negative economic consequences
for firms that rely on this market. Regulators face challenges when dealing with
stablecoins that are “too big to fail” and can rely on central bank support. Nonetheless,
the use of smart contracts to design more resilient financial structures may make
stablecoins “run-proof.”

Regulatory Concerns
As the world of stablecoins evolves, regulators face concerns similar to those faced by
the traditional banking industry. Banks create money by issuing deposit liabilities with a
fixed exchange rate against U.S. dollars, as well as credits held in Federal Reserve
accounts. They use low-earning liabilities to purchase high-earning assets. Commercial
banks are referred to as fractional reserve banks because they typically hold only a
small percentage of their assets as reserves. The implementation of federal deposit
insurance has largely eliminated retail bank runs. The Federal Reserve also provides
emergency lending to banks, but regulatory restrictions accompany these benefits on
bank balance sheets and business operations.

Some stablecoin issuers attempt to replicate the business models of banks or top
institutional money market funds, which can be a profitable venture. However, this
approach carries significant risks that could lead to instability. Other stablecoin issuers,
on the other hand, focus on providing payment services by holding only risk-free assets
such as U.S. Treasury securities. These stablecoins are similar to government money
market funds. Stablecoins may adhere to government regulations, hold only risk-free
assets, and generate profits via transaction fees or net interest margins. The goal is to
keep stablecoin users’ capital requirements low.

Advantages, Disadvantages, and Potential Applications

of a Central Bank Digital Currency
According to The Board of Governors of the Federal Reserve System (BOG), in its
recent publication “Money and Payments: The U.S. Dollar in the Age of Digital
Transformation,” a central bank digital currency (CBDC) is a “digital liability of the
Federal Reserve that is widely available to the general public.” In other words, this is
saying that anyone can open a bank account at the central bank.

Currently, only depository financial institutions and a limited number of agencies,

including the federal government, are allowed to have accounts at the Federal Reserve,
known as reserve accounts. These accounts hold bank reserves, and the account held
by the federal government is referred to as the Treasury General Account. Although a
form of CBDC already exists at a wholesale level for a select group of agencies, the
question of whether to expand access to it and how to do so remains. The public
already has access to digital currency in the form of digital deposit liabilities issued by
banks and physical currency, i.e., cash, which is considered a central bank liability.
However, bank deposits are not legally considered central bank or government
liabilities, whereas CBDC would be considered legal tender.

Federal Deposit Insurance

Bank accounts in the United States are today insured by the Federal Deposit Insurance
Corporation up to $250,000. It can be argued that retail bank deposits are a de facto
government liability. Furthermore, given the Federal Reserve’s role as lender of last
resort, large-value bank deposits can be argued to be a de facto government liability.
This means that the legal status of a CBDC in comparison to bank money may not be
as important in terms of money account security.

Counterparty Risk Concerns

When it comes to money and payments, security isn’t the only thing to think about.
There is also the issue of counterparty risk, which can impact access to funds. Even if
your money is insured in a bank account, accessing it may be delayed if the bank is
experiencing financial difficulties. This is one of the reasons corporate cash managers
frequently use the repo market, where deposits are secured by Treasury securities that
can easily be sold if the deposited cash is not returned on time. A CBDC with no
account size limits would provide fully insured money accounts for corporations with no
counterparty risk. If properly implemented, this could potentially help streamline certain
aspects of the money market.

Potential for Efficiency Gains

It’s a difficult issue to address when it comes to improving the overall efficiency of
CBDC’s payment system. In light of the current state of the U.S. payment system, which
lags behind developments in other countries, some argue that a properly designed
CBDC could be a game changer. However, it is important to note that the United States’
payment system is rapidly evolving, with platforms for real-time payment services, such
as the Federal Reserve’s FedNow and the Clearing House, emerging. It’s also important
to note that there is no “ideal” way to organize a payment system. A payment system’s
primary function is to process payment requests and transfer funds between accounts.
While the concept is simple, actual implementation and operation can be quite difficult.

In order to prevent fraud and ensure efficient messaging systems, any system would
require strong security measures. The potential of CBDC has divided opinion, with
some strongly in favor and others less so. Theoretically, a private, public, or
private-public partnership could all be viable options. The Federal Reserve is primarily
concerned with wholesale payments, whereas the private sector typically serves the
retail sector. A CBDC could be designed to respect this division of labor by allowing free
entry into the business of “narrow banking” or by providing direct access to CBDC and
delegating retail-level processing to private firms.

Question
Consider the fictional cryptocurrency, Plutocoin, a prominent blockchain-based system
that uses a unique blend of proof-of-work (PoW) and proof-of-stake (PoS) protocols.
Joe, a software engineer, is debating whether to use Plutocoin as his primary means of
transaction instead of conventional money and payment systems.
Which of the following would be the most accurate statement for Joe to make?

A. Plutocoin, being a cryptocurrency, does not provide the same security as

conventional banking systems.

B. Plutocoin transactions require the disclosure of personal information, similar to

conventional banking systems.

C. Plutocoin transactions are based on digital bearer instruments, making it akin to

digital cash.

D. Plutocoin transactions are only accessible to a certain group of people, unlike

conventional banking systems.

Solution

The correct answer is C.

Cryptocurrencies like Plutocoin are indeed akin to digital cash. They are digital bearer
instruments, where ownership control is defined by possession of a private key. Just as
physical cash doesn’t need permission to acquire and spend, cryptocurrencies can be
similarly managed. This is unique to cryptocurrencies compared to conventional
banking systems.

A is incorrect because Plutocoin, as a cryptocurrency, actually offers a high degree of

security, often more than conventional banking systems. This is due to the blockchain
technology and private key mechanisms that secure each account.

B is incorrect because one of the main features of cryptocurrencies like Plutocoin is the
ability to transact without having to disclose any personal information, unlike
conventional banking systems where personal information is usually required.

D is incorrect because, contrary to the statement, cryptocurrencies like Plutocoin can be

accessed by anyone with internet access and does not require any permissions or
qualifications to acquire or spend, unlike some conventional banking systems that may
have restrictions on who can open an account.
The Future Monetary
System
After completing this reading, you should be able to:

● Identify and describe the benefits and limitations of crypto and decentralized
finance (DeFi) innovations.
● Describe the role of stablecoins in DeFi ecosystems and differentiate among the
types of stablecoins.
● Discuss possible advantages and disadvantages of a monetary system based on
CBDCs.
● Understand the risks posed by the centralization that occurs in DeFi ecosystems
and crypto exchanges (CEX).
● Outline the regulatory actions recommended by the BIS to manage risks in the
crypto monetary system.

The monetary system is used by people all over the world to conduct payments and
other financial transactions. The organizations and structures that surround and support
monetary exchange make up the monetary system.

The central bank, which maintains the system’s fundamental functions and issues
money, is at the center of the monetary system. Trust in the central bank ultimately
underpins confidence in the monetary system.

The majority of payments are carried out by commercial banks and other private
payment service providers (PSPs), who also provide services to customers. This
separation of roles encourages competition.
Although the central bank-centered monetary system has done a good job historically,
new demands are being placed on it by technical advancements brought on by digital
innovations such as those in the crypto universe.

Decentralization is a foundational principle in the crypto universe. To make the system

self-sustaining and free from the influence of strong entities or groups, crypto envisions
checks and balances provided by several anonymous validators. Decentralized finance,
also known as “DeFi,” aims to replicate traditional financial services in the crypto
universe.

Recent occurrences have shown that the crypto ideal and its actuality are different.

● The collapse of the TerraUSD stablecoin and its companion coin Luna has
highlighted the system’s fragility.
● The crypto world, in particular, lacks a nominal anchor. It has become apparent
that crypto has deeper structural limits that prevent them from achieving the
effectiveness, stability, or integrity levels necessary for a suitable monetary
system.
● It is prone to fragmentation, and scaling its applications without sacrificing
security is impossible, as shown by the high fees.
● The system’s use of unregulated intermediaries is of grave concern.

Retail, fast payment systems (FPS) and central bank digital currencies (CBDCs), and
decentralization and permissioned distributed ledger technology (DLT) are well
positioned to serve the public interest by lowering costs and increasing convenience
while upholding the integrity of the system. Faster, safer, and less expensive payments
and settlements are supported by these developments domestically and internationally.

What do we Want from a Monetary System?

The monetary system, which comprises money and payment systems, is a collection of
institutions and structures that facilitate monetary exchange.

● Safety and stability – Money must do the following three functions in order to
guarantee the security and stability of the system: a store of value, a unit of
account, and a medium of exchange.
 ● Accountable – Key entities that are dependent on the financial system must be
held accountable, either through particular mandates for public authorities or
through appropriate regulation and oversight for private businesses.
● Efficient – To facilitate economic transactions on a large scale and at a cheap
cost, the monetary system should be effective and enable dependable, quick
payments.
● Financial inclusion – To spread the advantages of economic activity and
encourage financial inclusion, everyone should have access to affordable, basic
payment services, particularly transaction accounts.
● User control – The system must guarantee user control over financial data and
defend privacy as a fundamental right.
● Integrity – By preventing illegal activities such as fraud, financing of terrorism,
and money laundering, the integrity of the system must be preserved.
● Adaptable – The monetary system must be aware of technological
advancements, anticipate user needs and future advances, and adapt to shifting
demands from consumers and enterprises. Additionally, it ought to promote
innovation and competitiveness.
● Open – The monetary system also needs to be open, interoperable, and
adaptable both domestically and internationally in order to better serve a world
that is becoming more connected.

Challenges faced by today’s monetary system include:

● Users’ changing needs and concurrent technological changes have highlighted

areas for improvement.
● Current payment methods can occasionally be complicated and expensive to
use, which is partly due to a lack of competition.
● A sizable portion of individuals still lacks access to digital payment methods,
particularly in developing and emerging market economies.
● Demands for improvements to institutional structures and technological
advancements to meet the system’s changing societal demands.

The Promise and Pitfalls of Crypto

Stablecoins, which aim to tie their value to fiat currencies, highlight the crypto industry’s
desire to capitalize on the legitimacy offered by the unit of account issued by the central
bank. Stablecoins’ dependence on the legitimacy of central bank money is a major
structural weakness that is readily apparent. Additionally, stablecoins are frequently not
as stable as claimed by their issuers.

The abundance of coins also highlights another crucial weakness in the structure of
cryptocurrencies, namely their fragmentation into numerous incompatible settlement
levels.

The fragmentation of the cryptocurrency market reveals its challenge. The more users
that flock to a single blockchain system, the worse the congestion and the higher the
transaction fees, which invites newer competitors who might forgo security in favor of
greater capacity.

Crypto gives a sneak preview of potential functionality. These result from the ability to
bundle transactions and carry out the automatic settlement of bundled transactions in a
conditional manner, enabling higher functionality and speed.

The Building Blocks of Crypto

Bitcoin presented a revolutionary concept that gave rise to cryptocurrency: a
decentralized method of transferring value on a permissionless blockchain. The
validation of transactions on a public ledger can be done by any participant acting as a
validating node. Record-keeping on the blockchain is carried out by a variety of
anonymous, self-interested validators as opposed to depending on trusted
intermediaries (such as banks). Transactions with cryptocurrencies are validated by
decentralized validators and recorded on the public ledger.

How does Crypto Work?

A buyer broadcasts the specifics of the transaction when a seller wishes to send
cryptocurrency to them. Validators compete for the opportunity to verify the transaction,
and whoever is chosen to do so adds the transaction to the blockchain. The exact
names of the parties involved in transactions are therefore kept a secret, but the history
of every transaction is publicly viewable and linked to particular wallets.
In order to make the system self-sustaining, miners and validators receive financial
rewards for carrying out their duties in accordance with the rules. Cryptocurrency
rewards can be received in the form of transaction fees or from rentals earned by
“staking” one’s coins in a blockchain that uses proof-of-stake. The more frequently a
node acts as a validator, and the higher the rents are, the higher the stake.

Many blockchains and cryptocurrency coins have emerged since the creation of Bitcoin,
most notably Ethereum, which enables the use of “smart contracts” and
“programmability.” The use of self-executing code known as “smart contracts” can
automate market processes and do away with the middlemen who were previously
needed to make choices. Smart contracts are transparent and less susceptible to
manipulation because the underlying code is open to the public and can be examined.
The ability of smart contracts to integrate many parts of a system is a key aspect. By
merging several instructions into a single, smart contract, users can conduct
complicated transactions on the same blockchain. By “tokenizing,” they can turn assets
into digital representations.

Since smart contracts cannot directly access data that is “off-chain” or outside the
specific blockchain, they depend on intermediaries to supply this information (so-called
oracles). But these modifications also bring new issues. More centralized validation
techniques are frequently used to boost capacity, which weakens security and
concentrates insiders’ benefits.

Stablecoins in Search of a Nominal Anchor

With the stated goal of reshaping the financial system by eliminating middlemen and
lowering costs, decentralized finance provides financial services and products. Lending,
trading, and insurance are the three major forms of financial activity, though the DeFi
ecosystem is expanding quickly.

Lending. Users can make interest-bearing loans of their stablecoins to borrowers who
pledge other cryptocurrencies as security on lending sites.

Trading. Decentralized exchanges (DEXs) are venues for direct trades between holders
of cryptocurrencies or stablecoins.
Insurance. Users of DeFi insurance platforms can protect themselves against things like
improper handling of secret keys, exchange hacks, and unsuccessful smart contracts.

Due to their pegs to fiat money or other assets like gold, stablecoins are essential to the
DeFi ecosystem. They value the credibility that the asset or unit of account issued by
the central bank offers. Their primary use case is to combat the extreme price volatility
and limited liquidity of unbacked cryptocurrencies like Bitcoin. Their use also prevents
costly, frequent conversions between cryptocurrencies and bank deposits in fiat
currencies.

The two main types of stablecoin are asset-backed and algorithmic.

Asset-backed stablecoins like Binance USD and USD Coin are typically administered by
a central intermediary who invests the underlying collateral and organizes the
generation and redemption of the coins.

Algorithm stablecoins. To sustain their value in relation to the target currency or asset,
algorithmic stablecoins rely on sophisticated algorithms that automatically rebalance
supply.

The reliability of stablecoin stabilization methods heavily rests on the quality and
transparency of their reserve assets, both of which are frequently terribly deficient.

Stablecoins lack the characteristics required to support the upcoming monetary system.

● They cannot make use of the legal constraints and protections of bank deposits
or the central bank’s role as a lender of last resort.
● They must import their credibility from sovereign fiat currencies.
● They restrict liquidity and risk fracturing the monetary system.

Structural Limitations of Crypto

The inherent limits of permissionless blockchains, which eventually result in system
fragmentation, congestion, and excessive costs, are a problem for cryptocurrencies.
Transactions are recorded on the blockchain by self-interested validators.
In order to maintain the decentralized consensus system, validators must be motivated
by financial rewards that are sufficiently high. Sincere validation must generate more
rewards than possible gains from cheating. If compensation becomes insufficient,
validators may be tempted to cheat and compromise the security of the cryptocurrency.

The blockchain’s capacity must be restricted in order to maintain incentives for

validators and keep fees high enough. In times of congestion, users may offer larger
fees to have their transactions processed more quickly because validators have the
discretion to determine which transactions are approved and processed.

The so-called scalability trilemma is manifested in the restricted scale of blockchains.

Only two of the three properties—scalability, security, or decentralization—can be
attained by permissionless blockchains by virtue of their design. Decentralization and
incentives improve security, but maintaining incentives through fees causes congestion,
which restricts scalability. As more recent blockchains that compromise on security have
entered the crypto universe, the restricted scalability of blockchains has caused the
crypto world to become fractured.

There are increased dangers of theft and hacking in a system of rival blockchains that
are not interoperable but are supported by speculation. Interoperability refers to the
ability of users to access, share, and validate transactions across several blockchains.
Since each blockchain represents a distinct record of settlements, interoperability is not
possible in practice. To enable the movement of coins between blockchains, a few
“cross-chain bridges” have emerged. These bridges depend on a small number of
validators who, in the absence of regulation, must be believed to refrain from engaging
in unethical activity. However, a surge in hacks has coincided with the growth of these
bridges.

In contrast to the network effects that develop in conventional payment networks, the
crypto world is highly fragmented. In a conventional payment network, a platform gains
more users as more individuals use it. Trust and affordability are essential to such
platforms. Contrarily, cryptocurrency’s propensity for fragmentation and excessive fees
is a fundamental structural fault that disqualifies it from serving as the basis for the next
generation of money.
High price correlations between various cryptocurrencies and blockchains might be
caused by speculation. Increased demand drives up prices even further. Strong price
co-movement is present. When the rate of new users suddenly declines, there are
significant worries about what would happen to a system that depends on selling new
coins.

The DeFi Decentralization Illusion and the Role of

Exchanges
The DeFi ecosystem exhibits a propensity toward centralization despite its name. Voting
is used to make a number of important decisions among the owners of “governance
tokens,” which are frequently given to developer teams and early investors and are thus
highly centralized.

Since it is impossible to specify in contracts what to do in every scenario, conflicts must

be settled by some central bodies.

Additionally, aiming for increased throughput and speedier transactions, newer

blockchains typically rely on concentrated validation processes.

Incentives conflicts and hacking danger arise as validators become more centralized.

Furthermore, there are currently no regulations governing the screening of Oracle

providers, and anyone in control of Oracle can undermine the system by reporting
inaccurate data.

The cryptocurrency market is likewise centralized, with investors favoring centralized

exchanges (CEXs) over decentralized ones (DEXs). Similar to traditional banking, CEXs
keep off-chain records of the orders that traders have posted. Due to their cheaper
expenses, CEXs also draw greater trading activity than DEXs. Since 2020, CEXs have
experienced significant expansion and have attained quantities that qualify them as
important from the perspective of financial stability.

Before adopting relevant regulatory regulations, it is necessary to make a fair

assessment of the parallels and divergences between the crypto market and traditional
finance.
Regulatory Approaches to Crypto Risks
To address the immediate risks in the cryptocurrency monetary system, regulatory
action is required.

Authorities need to take regulatory arbitrage seriously. They should make sure that
crypto and DeFi operations adhere to the legal requirements for similar traditional
activities, operating under the guiding premise of “same activity, same risk, same
regulations.” The recent collapse of numerous stablecoins has brought attention to how
urgent the situation is.

To promote the integrity and safety of the financial and monetary systems, policies are
required. A fine or closure should be imposed on cryptocurrency exchanges that
conceal the identities of parties to transactions and disregard fundamental
know-your-customer (KYC) and other Financial Action Task Force (FATF) rules. They
could otherwise be used to dodge taxes, finance terrorists, or get around economic
restrictions. Similarly, banks and credit card firms need to demand user identity and
carry out KYC compliance.

Consumer protection laws. Investors ought to be able to invest in riskier securities, such
as cryptocurrencies, but there needs to be sufficient disclosure. This entails sound
regulation of the frequently deceptive advertising of digital assets. Front-running-like
behaviors can necessitate the application of innovative legal strategies.

The exposure of banks and nonbank financial intermediaries to the cryptocurrency

ecosystem poses dangers to the stability of the financial system that must be reduced
by central banks and regulators. Traditional financial institutions are investing in
cryptocurrencies at a rapid rate, which means that shocks to the cryptosystem might
have a ripple effect. Large traditional banks have so far only had minor exposures, and
their direct investments in businesses engaged in the cryptocurrency markets are still
modest when compared to their total capital. To address these risks, sound guidelines
for bank exposures to cryptocurrencies must be implemented.

It is crucial to discover regulatory entry points and close data gaps. A multitude of new
centralized intermediaries have popped up as a result of the cryptocurrency market’s
expansion. The traditional financial institutions and these centralized entities serve as a
natural entry point for regulatory responses.

The worldwide nature of crypto will necessitate collaboration across all regulatory fields.
In order to sanction non-compliant actors and platforms, authorities may need to
actively exchange information and conduct coordinated enforcement measures.

Crypto’s Lessons for the Monetary System

Overall, the cryptocurrency industry offers a glimpse of exciting technological potential,
but it is unable to achieve all the major objectives of a digital monetary system. It has
flaws such as stability, effectiveness, accountability, and integrity that legislation can
only partially resolve. Fundamentally, cryptocurrency and stablecoins result in a
fragmented and weak monetary system that comes from the economics of incentives
rather than technological limitations.

The goal is to harness the beneficial cycle of network effects by not only enabling
valuable functionalities like programmability, composability, and tokenization but also
putting them on a more secure base. Central banks are utilizing the best aspects of new
technology to serve the public good, along with their attempts to regulate the crypto
universe and address its most pressing issues.

Vision for the Future Monetary System

New technical advancements and a more accurate representation of central bank
money at its core should combine to create the future monetary system. The benefits of
new digital technologies can therefore be gained through interoperability and network
effects because they are rooted in trust in the currency.

As one of its primary functions is to issue central bank money, which acts as the basis of
the future monetary system and the unit of account in the economy, central banks are
ideally positioned to provide it.

Using its balance sheet, the central bank’s other function is to provide the means for the
ultimate finality of payments. It’s a trusted middleman who debits the account of the
ultimate payer and credits the account of the ultimate payee.
Another function of the central bank is to ensure enough liquidity available for
settlement to support the smooth operation of the payment system. L liquidity provision
ensures no bottlenecks will obstruct the smooth operation of the payment system.

Protecting the integrity of the payment system through regulation, supervision, and
oversight is the central bank’s other responsibility. The central bank is in a good position
to provide the groundwork for cutting-edge services in the private sector thanks to its
interconnected roles.

The future monetary system expands on these central bank functions and
responsibilities in order to provide room for new capabilities of central bank money and
cutting-edge services to be built on top of them. New private applications will be able to
function on more advanced technology representations of central bank money rather
than stablecoins. Thus, numerous new activities can be supported by innovations at the
central bank.

Components of the Future Monetary System

The future monetary system will be based on the tried-and-true division of labor
between the central bank, which supplies the framework of the system, and the private
sector organizations, which handle customer-facing operations. Additionally, new
standards can be created, like application programming interfaces (APIs), to
dramatically improve the interoperability of services and related network effects.

This vision includes elements at the wholesale and retail levels that make a variety of
new features possible. At the wholesale level:

Programmability. It can facilitate transactions between financial intermediaries and

provide new capabilities. Using permissioned distributed ledger technology (DLT),
CBDC transactions allow programmability and settlement, enabling transactions to be
conducted automatically when predetermined conditions are met.

Composability. make it easier for transactions to be composed by enabling the

combination and execution of a variety of different functions.
A wider range of financial intermediaries. New capabilities not only enable a far larger
range of financial intermediaries to participate in transactions but also allow for an
increase in the types of transactions.

Multi-CBDC arrangements. Additionally, Central Bank Digital Currencies (CBDCs)

collaborate internationally through multi-CBDC agreements, including various central
banks and currencies.

Tokenization. When deposits are tokenized, they are decentralized, settled, and given a
digital representation on the distributed ledger technology (DLT) platform. This might
make it easier to exchange new technologies, including fractional ownership of stocks
and other assets, opening the door to creative financial services that go far beyond
simple payment processing.

At the retail level:

Interoperability. Users of one platform can send messages and commands to others
that are linked to it with ease through APIs. Retail innovations encourage more
competition, decreased costs, and increased financial inclusion in this way.

Instant payments. Households and businesses can easily access retail CBDCs.
Through various interfaces and competing private Payment Service Providers, they
enable immediate payments between end users.

Data architecture. The Central Bank Digital Currencies (CBDCs) are backed by a data
architecture that includes digital identity and APIs that allow for safe data interchange,
enabling more user control over financial data.

Open platform. They facilitate lower costs in payment services by fostering efficiency
and better competition among private sector PSPs through the provision of an open
platform.

Inclusive design. Both can help financial inclusion for people who do not already have
access to digital payments through inclusive design elements.

A Metaphor for the Future Monetary System

The central bank serves as the solid base of the metaphorical tree that represents the
future monetary system. The tree metaphor illustrates the idea that the monetary
system is founded on payment finality through eventual settlement on the central bank’s
balance sheet.

The central bank-based monetary system fosters a thriving ecosystem of participants

and activities where rival private sector Payment Service Providers can employ
creativity and innovation to better serve users.

When viewed from a distance, the world’s monetary system can be compared to a
forest, whose canopy makes cross-border and cross-currency activities possible.
Infrastructures like multi-CBDC (Central Bank Digital Currencies) platforms function as
crucial new system components in the canopy.

Innovation is crucial. For private innovation to serve the public good, the fundamentals
must be done correctly. Public infrastructures can enhance the payment system by
utilizing many of the purported advantages of cryptocurrencies without the downsides,
as demonstrated by ongoing work at central banks. How central banks may assist
interoperability and data governance is demonstrated through wholesale and retail
CBDCs as well as open banking initiatives.

Wholesale CBDCs and Tokenized Money

A Central Bank Digital Currency (CBDC) is a digital payment instrument that is directly
owed to the central bank and is valued in the country’s unit of account. Retail CBDCs
are available to both individuals and companies. However, unlike domestic commercial
banks, wholesale CBDCs provide additional payment and settlement capabilities to a
considerably wider range of intermediaries, potentially spurring innovation.

Multi-CBDC systems involving several central banks and currencies can benefit from
decentralized governance. Functions such as self-executing smart contracts are
essential because they enable users to limit the settlement of their transactions to the
event that certain predefined conditions are met (programmable). Through such
automation, payments, and delivery of securities can be done only all at once or not at
all, accelerating settlement and reducing counterparty risk.
Wholesale CBDCs have the advantage of potentially being accessible to a far broader
range of intermediaries than domestic commercial banks. By allowing nonbank PSPs to
conduct CBDC transactions, competition, and vitality could increase significantly.

Payments could be finalized with wholesale CBDCs. In a digital system, establishing the
source or “provenance” of the transferred funds is necessary to ensure that payment is
genuine. Cryptography is used to conceal real names while openly displaying the
complete history of all transactions made by all parties.

The tokenization of regulated financial instruments, including retail deposits, can be

supported by central banks. Commercial bank deposits can be represented digitally
through tokenized deposits. Tokenized deposits would be programmable, “always on”
(24/7), and covered by deposit insurance, making them suitable for a wider range of
retail payment applications, such as autonomous ecosystems. This approach could
make it easier for other financial assets, like stocks or bonds, to be tokenized, enabling
fractional ownership of those assets.

One potential tokenized deposit system would include a permissioned platform that
keeps track of every transaction in tokens produced by the participating institutions.
Retail investors (depositors) would store tokens in digital wallets and send tokens
between wallets to make payments. This might take place concurrently as a result of
smart contract-enabled single atomic transactions. Digital representations of bonds and
stocks might potentially be possible with the same system.

In autonomous ecosystems, programmable CBDCs might also facilitate

machine-to-machine payments. Machines can manage their own budgets and make
direct purchases from one another for goods and services. Their integration will
enhance the need for programmable money and smart contracts, lowering any
settlement risk. This might result in huge efficiency savings, for instance, in the goods
logistics industry, where transactions are still primarily paper-based and can take
several days.

The advantages of atomic settlement and open-source protocols can coexist peacefully,
with central banks serving as the main hubs for validation.
Retail CBDCs and Fast Payment Systems
Wholesale and retail CBDCs have a lot in common. Retail CBDCs provide consumers
and companies with digital access to money from the central bank. Retail-facing
payment services are provided by PSPs, both bank and nonbank. Thus, retail CBDCs
are occasionally referred to as “digital cash.” As transfers take place in real-time or
close to real-time, on an almost 24/7 basis, they have great speeds and availability.

Contrary to crypto, which needs high rents, has limited scalability, and is congested,
Central Bank Digital Cryptos and retail Fast Payment Systems (FPS) provide broader
use, which lowers costs and results in better services.

Retail Fast Payment Systems (FPS) have already made notable strides toward lowering
costs and promoting financial inclusion for the unbanked, particularly in Brazil, where
two-thirds of the adult population use their retail FPS Pix.

Retail CBDCs can be created to assist financial inclusion by examining retail CBDC
design elements that address particular barriers to financial inclusion, such as through
novel interfaces and offline payments. The cost of payment services for the unbanked
can be decreased by central banks through tiered CBDC wallets with simpler due
diligence for users transacting in lower values, promoting more access to digital
payments and financial services.

As every transaction in the digital economy leaves a trace, generating issues with
privacy, data misuse, and personal safety, retail Central Bank Digital Cryptos (CBDCs)
and Fast Payment Systems (FPSs) can be created with privacy protection and
improved user control over data in mind. Since there isn’t a single, comprehensive
record of every transaction in conventional payment systems, individual Payment
Service Providers (PSPs) have more control over data. Every PSP, however, only
records its own transactions. Even the central operator is unaware of all payments in
full. Thus, the promise of confidentiality by the central operator and the mix of
segregated record-keeping are used to protect the privacy of payments.

Crypto advocates claim permissionless blockchains give consumers back control over
their personal data, yet this poses serious threats to users’ privacy and integrity. Retail
FPSs and CBDCs’ underlying structures can allow users control over their personal
data while protecting their privacy and promoting consumer welfare. Additionally, central
banks can genuinely build systems in the public interest because they have no financial
interest in personal data.

Systems that provide law enforcement authorities with access to information while
maintaining the necessary legal protections can be created with clear mandates and
public accountability. These methods are currently being considered for retail CBDCs
and are already standard practice in the form of bank secrecy rules.

Retail CBDCs and FPS have the potential to enhance accountability in comparison to
the current system and the crypto universe. The role of the central bank in retail
payments needs to be discussed publicly as a result of new systems, and legal
regulations will need to be modified in order for them to operate. Private service
providers would be integrated into a strong regulatory and oversight framework under a
system based on public infrastructure.

Question
Jessie, a risk manager at a venture capital firm, is evaluating potential investments in
crypto and decentralized finance (DeFi) start-ups. She recognizes that these
technological innovations can revolutionize the financial industry. However, she must
also take into account their potential risks and limitations. As part of her analysis, she is
considering the structural integrity and robustness of the technologies powering these
potential investments.

From the following choices, which statement most accurately captures Jessie’s
concerns about the possible limitations of crypto and DeFi technologies?

A. The presence of self-interested validators in a crypto system automatically ensures

its safety and integrity.

B. The limited scalability of blockchains is an indicator of their secure nature, making

them ideal for all financial transactions.
C. The fragmentation of the crypto universe, combined with the so-called scalability
trilemma, underpins inherent limitations in blockchain systems.

D. The “cross-chain bridges” between different blockchains guarantee interoperability

and completely eliminate the risks of hacking and theft.

Solution

The correct answer is C.

The fragmentation of the crypto universe, combined with the so-called scalability
trilemma, underpins inherent limitations in blockchain systems. This is because the
proliferation of different cryptocurrencies (fragmentation) could reduce overall efficiency
and compromise interoperability. The scalability trilemma underscores that blockchain
systems can only optimize two out of the three fundamental properties: security,
scalability, and decentralization. Hence, a system may offer excellent security and
decentralization but face challenges in scaling up to accommodate large numbers of
transactions. These combined issues present significant potential limitations of
blockchain technology.

A is incorrect as the presence of self-interested validators does not automatically ensure

the safety and integrity of a crypto system. Validators in a pseudo-anonymous crypto
system have no reputation at stake and cannot be held accountable under the law, and
must be incentivized through monetary rewards to maintain the system’s integrity.

B is incorrect as the limited scalability of blockchains, while enhancing security to some

extent, is not ideal for all financial transactions. The scalability limitations result in
system fragmentation and high transaction fees, posing significant challenges for
wide-scale adoption.

D is incorrect as “cross-chain bridges” do not completely eliminate the risks of hacking

and theft. These bridges often rely on a small number of validators, and their use has
been associated with several high-profile hacks due to weaknesses in governance.
Climate-Related Financial
Risks – Measurement
Methodologies
After completing this reading, you should be able to:

● Describe the main issues in identifying and measuring climate-related financial

risks.
● Identify unique data needs inherent in the climate-related risks and describe
candidate methodologies that could be used to analyze these types of data.
● Describe current and developing methodologies for measuring climate-related
financial risks employed by banks and supervisors.
● Compare and contrast climate-measuring methodologies utilized by banks,
regulators, and third-party providers.
● Identify strengths and weaknesses of the main types of measurement
approaches.
● Assess gaps and challenges in designing a modeling framework to capture
climate-related financial risk.

Main Issues in Identifying and Measuring Climate-related

Financial Risks
In this section, we discuss general issues in measuring climate-related financial risks
and the translation of underlying concepts to concrete climate risk measurement.
Climate change may cause both economic and financial effects, which may lead to
losses for banks. As a result, banks should have a risk management framework that
identifies and measures climate risk drivers, maps and measures climate-related
exposures, identifies areas of risk concentration, and converts climate-related risks into
quantifiable financial risk metrics.

An overview of measurement methodologies that banks and supervisors are currently

applying is discussed here based on conceptual issues related to climate-related
financial risk measurement.

Methodological Considerations
According to BCBS (2015), effective risk governance involves mapping and measuring
risk exposures. The assessment of climate-related financial risks is similar to that of any
other risk. The only difference is that climate-related financial risks have unique features
that challenge traditional risk measurement.

Conceptual Considerations in Measuring Climate-related Financial Risks

As risk managers assess climate-related financial risks, they learn new concepts. These
concepts are heavily used in mapping and measuring climate-related financial risks.
The following are some key concepts supervisors and banks need to know:

Physical and Transition Risk Measurement Considerations

Climate-related financial risks have several unique characteristics. Physical risks and
transition risk drivers drive these characteristics, resulting in different exposure mapping
and measurement approaches.

Generally, physical risk drivers (physical hazards) can be associated with financial
exposures using damage functions that show how a specific risk driver affects tangible
assets. We can attribute disruptions to assets, activities, and related financial flows to
comprehensive risk models that measure financial risk parameters.

The damage function applied within a specific risk model depends on a bank’s
technological and resource capacity, the availability of relevant data, and the purpose
for which the estimation is intended. In addition, sectors, the severity of hazards, time
horizon factors, and geospatial idiosyncrasies determine specific damage functions.
As a result of their distinct characteristics, physical and transition risks are assessed
separately. Some climate change features, however, increase the probability of
dependence among these risks. Hence, they can be viewed jointly.

Exposure Granularity
Bank’s transactions with other parties may expose it to climate-related financial risks.
Banks and supervisors will have to decide the level of granularity at which to assess the
implications of climate risk drivers for these transactions. The following factors may
influence this decision:

● Specific physical or transition risk drivers.

● Availability of relevant data.
● Risk management decisions being supported.
● Increasing computational complexity with increasing granularity.

This decision drives the selection of available approaches, which then affects a model’s
output for risk management.

Top-down and Bottom-up Approaches

The selection of top-down or bottom-up is among the conceptual considerations that
apply to both exposure mapping and risk estimation.

Top-down approaches typically start by estimating risks at a general or aggregated level

and then allocate (push down) the risk measure to individual parts.

On the other hand, bottom-up starts by dimensioning risk at the component level and
then aggregating these individual risks to provide a general view of risk.

Incorporating Risk Mitigation and Risk Reduction

Banks may need to estimate how potential risk mitigation might moderate or offset
climate-related financial risks when considering how to measure them. Climate risk
measurement approaches that incorporate mitigation strategies are thought of as
showing net exposure. On the other hand, approaches that do not incorporate offsetting
strategies are viewed as showing gross exposure. Banks can disaggregate the impact
of risks and mitigating actions by distinguishing between net and gross exposures.

Climate-related financial risks can also be offset through counterparty measures to

adapt to or mitigate the effects of climate change. The relationship of exposures to risk
drivers within the risk models of the bank may be modified by these measures. We have
two reasons why we should calculate the gross exposure of an asset or portfolio:
 1. By understanding exposure without incorporating risk-offsetting practices, risk
decision-makers will be able to assess the current magnitude of climate-related
risks as well as the future evolution of these risks.
2. Mitigants may lapse, change, fail to materialize, or become obsolete, reducing
their effectiveness in reducing risk.

Heterogeneities
Heterogeneity is an important consideration when selecting measurement approaches.
Climate-related financial risks vary by bank, depending on geographical location,
markets, sectors, political environments, and technology. The following are among the
common heterogeneities:

● Geographical heterogeneity: Climate hazards may vary according to

geographical location.
● Sectoral heterogeneity: Corporate counterparties tend to have similar exposures
to transition and physical risks.
● Jurisdictional heterogeneity: Borders of a particular nation define the limits of
legal. jurisdiction. Exposures within the same jurisdiction may be subject to
varying policies.

Sources of Uncertainty
There is uncertainty associated with estimating climate-related financial risks. As new
information is incorporated into climate models, climate sensitivity estimates typically
trend higher, suggesting an underestimation of climate-related financial risks.

Unique Data Needs Inherent in Climate-related Risks

Unlike traditional data that banks normally use in financial risk analyses, assessing
climate-related financial risks will require new and unique data types.

We have three broad data categories needed to assess climate-related financial risk:

1. Physical and transition risk drivers data: These are needed to translate climate
risk drivers into economic risk factors (i.e., climate-adjusted economic risk
factors).
 2. Vulnerability of exposures data: These link climate-adjusted economic risk factors
to exposures.
3. Financial exposure data: These are required to translate climate-adjusted
economic risk into financial risk.

1. Physical and Transition Risk Drivers Data

This data type is the foundation for assessing the effects of climate-related risks on
banking exposure. Climate data and hazard event data belong to this data category. We
can use these data as independent variables in order to alter existing economic
relationships and influence economic outcomes. Government agencies and academic
institutions usually provide this data type, while commercial third parties supply some.

2. Vulnerability of Exposures Data

Besides climate-adjusted economic risk factors, banks and supervisors need
information about the vulnerability of bank exposures to those risk factors. These data
tend to include features specific to those exposures, such as geospatial data for
corporates, location data for mortgage collateral, or data on the sensitivity of
counterparties to energy prices.

In addition to being used in measurement approaches, these data facilitate the

translation of climate-adjusted economic risk factors into financial exposures. Generally,
the relevant characteristics of these data differ according to the climate risk driver under
consideration.

Climate risk drivers determine the relevant characteristics of these data. Counterparties’
geospatial location primarily determines physical risk exposure. On the other hand, the
vulnerability of bank exposures to transition risk depends, however, on the economic
activity of counterparties within particular jurisdictions.

Physical risks can result in the destruction of property and inventory, which can affect
the economy. As a result, physical hazards should be sufficiently matched with the
location of relevant physical touchpoints in order to assess the vulnerability of exposure
to physical risks.
It may also be necessary to collect information about interconnections between retail,
corporate, and municipal borrowers in order to evaluate the impacts of deteriorating
local economic conditions due to a severe or chronic weather event on the local
economy. To assess the vulnerability of corporate counterparties’ exposure to transition
risk, data concerning their sectors and subsectors, as well as their carbon sensitivity,
are often needed.

3. Financial Exposure Data

In order to translate the vulnerability of exposures into financial loss estimates,
additional data is usually needed. Banks can turn to variables used in conventional risk
measurement approaches. Such data include data used in estimating cash flows,
valuations, or prices. Data on portfolio composition and counterparties (e.g., probability
of default and loss given default) is required to analyze banks’ risk. In addition, data on
rollovers, withdrawals, or pricing will be needed to model potential bank liquidity impacts
due to climate risk.

Potential Methodologies
This section discusses conceptual modeling and risk measurement approaches that can
be used to estimate climate-related financial risks.

Integrated assessment models (IAMs) combine energy and climate modeling

approaches with economic growth modeling. AIMs link projections of transition risk
drivers and greenhouse gas (GHG) emissions to economic growth impacts. Even
though IAMs have been in use for quite some time, they fail to capture the economic
impacts of climate change, extreme weather events, and adaptation possibilities.
Furthermore, estimates of total GDP losses from changes in climate produced by IAMs
may not be realistic since only some physical risks are considered. Climate models
which underpin IAM projections may underestimate the severity of future outcomes if
they fail to capture physical risks which are clearly understood or measured.

Input-output models estimate how shocks will affect a sector or region’s economy based
on static economic linkages among sectors and geographical locations. In the context of
climate economics, an input-output accounting framework is used to examine the
impact of policy changes such as an emissions tax or to estimate the supply chain
impacts of extreme climate events.

Computable general equilibrium (CGE) models allow policy experiments with complex
behavioral interactions between sectors and agents that cannot be solved analytically
due to their complexity. Even though some mechanisms can be explained for CGE
outcomes, the level of complexity makes it impossible to assess the overall significance
of each embedded decision rule and parameter value governing economic agents,
resulting in a significant black-box effect.

Dynamic stochastic general equilibrium (DSGE) models make macroeconomic

modeling even more complicated, especially uncertainty in agent decision-making and
endogenous technological change. These models involve complex computations that
can be difficult to solve. Some academic institutions and central banks are taking the
necessary steps to make these models more useful in estimating the impacts of
climate-related risks.

Overlapping generation (OLG) models are a more transparent and stylized approach to
analyzing long-term macroeconomic evolutions. By examining the intergenerational
distribution of consumption, these models can highlight one key shortcoming of other
approaches: the large role discount rates play in estimating the social cost of carbon.

Agent-based models (ABMs) are the most recently introduced model for measuring
climate-related impacts due to their ability to reflect uncertainty and complexity better.
An ABM is a simulation in which economic actors interact with institutions and each
other based on a set of decision-making rules imposed by the modeler.

Other broad risk measurement approaches currently being used by banks and
supervisors include risk scores, scenario analysis, stress testing, and sensitivity
analysis.

Climate risk scores rate the climate risk exposure of assets, companies, portfolios, or
even countries. In order to assign a quality score to exposures, they combine a risk
classification scheme with a set of grading criteria. Banks and supervisors can use
climate risk scores to assess the relative climate exposure of existing and take
necessary credit interventions.

Scenario analysis can help quantify tail risks and clarify climate-related uncertainties by
examining a wide range of plausible scenarios. Climate scenario analysis involves four
steps:

1. Identify physical and transition risk scenarios.

2. Relate the effects of scenarios to financial risks.
3. Assess the sensitivities of the counterparty and sector to those risks.
4. Generalize the effects of these sensitivities to find an aggregate measure of
potential losses.

Stress testing is a subset of scenario analysis aimed at evaluating a bank’s near-term

resilience to economic shocks, often through a capital adequacy target.

Sensitivity analysis is another subset of scenario analysis that examines the impact of a
specific variable on economic outcomes. Typically, one parameter is altered across
multiple scenarios, and outputs are observed to see what happens each time the
parameter is altered.

Among other new approaches developed, we have natural capital analysis and climate
value-at-risk.

Natural capital analysis evaluates the negative effects of degradation on a bank. This
process typically involves four steps:

1. Identifying relevant geographies, sectors, borrowers, and/or assets.

2. Identifying natural capital assets (water, clean air, forests).
3. Identifying potential natural disruptions.
4. Identifying areas, sectors, borrowers, and assets most at risk.

Climate value-at-risk (VaR): This approach involves assessments applying the

traditional VaR framework to gauge the impacts of climate change on banks’ balance
sheets.
Measurement Methodologies of Climate-related Financial
Risks
In this section, we discuss methodologies for mapping and measuring climate-related
financial risks as well as their strengths and weaknesses. We also discuss scenario
analysis, stress testing, and sensitivity analysis, which are used to quantify
climate-related financial risks. Measurement methodologies are discussed separately
for banks, supervisors, and third parties, even though we may have similarities.

Exposure Mapping and Measurement

Bank-level Methodologies
Portfolio and sectoral exposures: Measures of carbon/emission intensity, energy
efficiency or energy label distribution of a real estate, or physical risk vulnerability of
collateral positioned in risky regions are among the examples here. Banks have
launched internal processes to evaluate climate-related financial risks qualitatively and
map their potential impacts in the absence of quantitative information. The indicators or
metrics that banks employ to map, measure, and monitor their exposures can either be
distinguished by transition risk or physical risk.

Transition Risks
The following are observed practices among banks:

● Through assessing the possible sources of shocks and transmission

mechanisms, banks usually analyze reasons for and the extent to which a
particular sector could be impacted by a transition to a low-carbon economy.
● As a proxy for transition risk, banks calculate the carbon footprint of their assets.
● Indicators of “greenness” of real estate and financial assets are proxies for
transition risk, as well as measures of alignment with climate targets. Portfolio
alignment involves measuring the gap between existing portfolios and a portfolio
that meets a specific climate target.
● Analyzing the potential risk differential between “green” and “brown” activities is a
backward-looking analysis.

Physical Risk
 ● Indicators metrics used to map, measure, and monitor physical risk are intended
to identify concentrations of risk geographically and their type, probability, and
severity.
● Location-based physical risk scores have been developed for physical risk
drivers such as heat stress, wildfires, floods, and sea level rise. Indicators are
identified for each risk driver that captures changes in physical conditions. A
bank can aggregate these indicators and translate them into facility-level scores,
which can then be used to perform due diligence on clients or to evaluate
mortgages annually.
● In addition, some banks are using geospatial mapping to assess and monitor
how physical risks may affect their exposures.

Client or project Ratings and Scores

● It is becoming increasingly common for large banks to assess climate-related
financial risks to which individual counterparties may be exposed to both physical
risk and transition risks. Banks have started integrating project-related climate
risk assessments into credit management procedures.
● As part of a counterparty-based risk assessment, banks evaluate the
climate-related opportunities and risks of the companies they finance or are
considering financing.
● Banks often use rating or scoring approaches. Ratings can be internal or
external. Climate-related risks are commonly assessed separately from standard
credit risk assessments.
● Rating metrics can be based on sector-level characteristics and then adjusted to
fit company-specific characteristics.
● To date, banks have integrated climate risk ratings into their overall customer
credit rating only on targeted occasions, although climate risk ratings generally
reflect climate-related factors when granting credit. Climate-related ratings are
being developed by several banks as part of their counterparty credit assessment
processes.

Supervisory Methodologies
Supervisors use similar metrics and indicators that banks use to map, measure, and
monitor exposure to climate-related financial risk.

Transition Risks
● The impact of transition risks is often assessed on the basis of regulatory
information or ad hoc surveys.
● Bank supervisors typically use indicators for the emissions intensity, carbon
footprint, or sensitivity to climate policies of banks’ counterparties at the entity or
sectoral level for corporate portfolios, except for real estate exposures.
● We only have a few analyses for real estate exposure transition risks.

Physical Risks
● Even though the analysis of exposures to physical risks varies and its
assessments are often in their initial stage, some supervisors have managed to
identify hazards, map and measure exposures.
● Supervisors identify hazards relevant in their jurisdiction and specific regions
more vulnerable to them. Information from third parties is crucial in conducting
the above analysis. Such information includes publicly available information and
climate risk scores provided by commercial vendors.
● The use of these indicators varies in terms of hazard, geographical granularity,
and complexity. Use of flood maps, country vulnerability indicators, and individual
industrial facilities’ water stress levels
● The authorities may assess the risk exposure of individual supervised entities or
of the banking system to geographies that are more susceptible to physical risk
once salient physical risk drivers have been identified.

Risk Quantification: Scenario Analysis, Stress Testing, and Sensitivity

Analysis
Bank-level Methodologies
Stress testing or scenario analysis methods can be used to quantify climate-related
financial risks at the bank level.

● Most of these exploratory exercises focus on credit risk or market risk analysis.
Climate-related financial risk scenario analyses are currently being conducted to
understand potential impacts on selected portfolios, refine methodologies, and
 assess limitations and benefits. Such exercises are used to build capacity and
identify counterparties that need to be involved in the transition process.
● Banks’ transition risk analysis focuses on the impact on credit parameters for
specific sectors of counterparties. Specifically, this includes corporate exposure
to climate policies in relevant sectors.
● Using a “shadow price” as part of a transition risk sensitivity exercise or scenario
is an example. This approach involves defining a range of possible future prices
on the basis of external scenarios or internal expertise. We can then use these
“shadow prices” to evaluate the effects on financial variables and hence the
counterparty’s credit risk profile.
● Banks’ physical risk analysis focuses on corporate and household exposures
(particularly mortgages). Based on this, they infer a potential impact on
counterparty credit quality. Companies in specific sectors (such as electric
utilities) can also be affected by business interruption assumptions.
● Observed practices include focusing on sectors more sensitive to weather
patterns.

Supervisory Methodologies
● Supervisors employ scenario analysis and climate stress tests for
micro-prudential supervision and for informing macro-prudential policies.
● At the micro-prudential level, scenario analysis, and stress testing may be used
to: (i) quantify banks’ financial exposures vulnerable to specific climate risk
drivers and/or (ii) understand the vulnerability of banks’ business models under
specific climate scenarios.
● At the macro-prudential level, scenario analysis, and stress testing can be
employed to assess whether climate risks are systemic in nature and determine
their size and distribution.

Third-party Approaches
Supervisors and banks sometimes rely on comprehensive methodologies or tools
provided by third parties in addition to specific data or metrics. The same features apply
to third-party methodologies as in the context of banks and supervisors, such as
exposure mapping, scenario selection, shock introduction, and impact assessment.
 ● A variety of risk metrics or tools may be provided by the methodologies, including
climate VaR, PD, expected shortfalls, expected losses (EL), income predictions,
revenue/cost analysis, LTV ratios, return on equity, and climate targets.
● For physical risks, a risk indicator is often proposed in the form of a climate risk
score. Some methodologies assign a risk rating to an exposure based on the
type of hazard to which it is exposed and vulnerable. Other methodologies leave
room for different types of segmentation or aggregation strategies.
● Methodologies developed by third parties incorporate climate sensitivity or
vulnerability, such as water and energy resources, or analyze the types of
facilities that are at risk. Several methodologies consider alternative production
sources and insurance policies as well. Vulnerability is assessed by using
damage functions that describe (historical) relationships between natural disaster
magnitude and asset damage.

Strengths and Weaknesses of the Main Types of

Measurement Approaches
The table below summarizes the strengths and weaknesses of the main types of
measurement approaches.
Gaps and Challenges in Designing a Modeling
Framework to Capture Climate-Related Financial Risk
In this section, we discuss key challenges in designing a modeling framework.

Limitations of Observed Risk Classification Approaches

Aggregate risk classification approaches not only have advantages but also face
several limitations. As a result of the current availability of data, identification criteria
may not be sufficiently granular to distinguish counterparties. For example, some risk
classification approaches assume that counterparties from the same geographical
location have the same risk characteristics. In reality, however, the transition and
adaptation capabilities of counterparties may vary depending on the specific physical or
transition risk driver being assessed.

In addition, the sensitivity of an individual counterparty to climate-related risks may not

necessarily be reflected in a sector or geographical area that is transition-sensitive.

It may be difficult to distinguish between gross and net exposure without being able to
clearly identify and measure the hedging strategies employed by counterparties.

Risk Differentiation and Comparability Across Banks and Jurisdictions

While risk classification systems strive for comparability, it comes at a cost. In order to
facilitate comparisons between banks’ exposures within or across jurisdictions,
standardization and/or simplification elements are introduced. So, there is a trade-off
between the level of detail and complexity needed to accurately assess risks and the
need to compare and combine risk information from different banks.

Challenges in the Availability of Appropriate Data

Recently, the availability of data and information has been reported as a major
challenge hindering the development of climate risk measurement processes.

Data Describing Physical and Transition Risk Drivers

A major challenge here is that information may be out of the scope of traditional
financial data collection. The available information may lack sufficient granularity and
further drawbacks that hinder the analysis of the data.

Data Describing the Vulnerability of Exposures

Third-party Rating Information
Ratings provided by third parties may not be fully reliable as the users of the ratings
may not be so sure about the accuracy of the data provided by the rated firms. Data
users may find it difficult to identify the methodological approach adopted by data
providers since most scores are based on proprietary models. In addition, the
comparability of indicators across vendors is also limited. Therefore, reconciling these
approaches is often difficult.

Counterparty-level Information
Counterparties provide lenders with proprietary non-public information in order to
develop a banking relationship and assess the counterparty’s creditworthiness. By
acquiring non-public client information via the lending relationship, the bank is able to
address some of the data gaps or quality issues on a bilateral basis. However, for small
counterparties, the availability of proprietary climate-related client data may be
qualitative rather than quantitative. Consequently, data completeness and precision
issues may materialize. Furthermore, banks rarely update data after the underwriting
processes are done. This could create gaps in climate reporting for existing exposures.

When proprietary non-public information is not readily available, banks may decide to
use public information disclosed by borrowers. It is worth noting that the quantity and
quality of public information depend on the size of the firm. Consequently, it limits the
comparability of smaller firms with large corporations.
Supervisory Reporting Data
Supervisory reports provide recurring and standardized data useful for measuring
climate risk. Their analysis could provide macro- and micro-level insights into banks’
asset portfolios. While existing data can be leveraged in combination with third-party
providers, current supervisory reporting may not provide sufficient granularity to assess
transition and physical risks.

Challenges in Designing a Modeling Framework to Capture Climate-related

Financial Risks
Scenario Design and Complexity of Climate-related Financial Risks
Financial losses are computed by incorporating all necessary economic and financial
variables at a granularity consistent with a risk classification. However, climate-related
financial risks are complex and coupled with uncertain climate drivers that go beyond
intrinsic future uncertainty inherent in physical and transition risk drivers. Banks and
supervisors cite the following aspects as being particularly challenging in modeling
comprehensive scenarios:

● Uncertainty around the climate risk drivers involving both transition and physical
risk drivers hinders measurements of climate-related risks.
● Capturing the specific impacts of climate scenarios is still a challenge, and hence
there is a need to continue developing modeling frameworks to capture the
impacts of climate scenarios, including stressed variants, within an integrated
and tractable modeling framework.
● There can be issues with the comprehensiveness of modeled impacts. In
particular, capturing the impact of climate scenarios at a level of granularity
consistent with a risk classification is challenging.

Translating Scenario Outputs to Financial Risks

It is particularly challenging to factor in climate scenario-related variables. Models
typically use historical statistics to estimate the impact of given risk drivers on credit risk
parameters, such as PDs or LGDs. To estimate robust relationships, historical
observations of risk drivers must have sufficient depth and variance. There are no past
observations of climate-related risk materializations that can be used to predict future
trends. Moreover, financial models cannot generate empirical risk parameters.
Time Horizon-related Challenges
Banks and supervisors have to consider horizons longer than usual due to the long-term
nature of climate change, which makes stress testing exercises and forward-looking
assessments challenging.

In addition, uncertainties associated with climate sensitivity modeling are exacerbated

by such long-term horizons, which also impact economic and financial projections,
thereby limiting their reliability when assessing risks. In addition, the uncertainty of a
model’s projections is directly proportional to the length of the horizons used in
forecasts. Limitations in modeling will further reduce projection robustness.

Banks have limited ability to internalize negative feedback associated with their
short-term lending decisions. To capture risks over longer time horizons, banks will need
to examine the appropriateness of existing measurement approaches as well as
possible modifications.

Operational Complexity in the Measurement of Risk

Measurement of climate risk requires adequate infrastructure, relevant human
resources, and/or sophisticated organizations. In order to assess their overall exposure
to climate risks across all their significant operations, banks need to be able to
aggregate and manage large amounts of data. The ability to collect, format, and
process enormous amounts of climate-specific data is vital to most methodologies.

In order to measure climate-related financial risk, it may be necessary to pool resources

from various business and functional areas, as well as to develop climate-specific
expertise in-house or hire experts externally. A bank’s size and complexity also
influence its choice of risk measurement methodologies. Internal harmonization towards
common risk assessment approaches, metrics, and methodologies may be challenged
due to idiosyncrasies among business lines and banking entities. Smaller, less complex
banking groups may face sophistication and resource allocation trade-offs.
Artificial Intelligence Risk
& Governance
After completing this reading, you should be able to:

● Identify and discuss the categories of potential risks associated with the use of AI
by financial firms and describe the risks that are considered under each category.
● Describe the four core components of AI governance and recommended
practices related to each.
● Explain how issues related to interpretability and discrimination can arise from
the use of AI by financial firms.
● Describe practices financial firms can adopt to mitigate AI risks.

AI risks refer to the potential ills that may cause harm to organizations, consumers, or
society at large. Such risks may originate from:

● Data used to train the AI system.

● Usage of the AI system.
● Poor governance of the system.

According to the Artificial Intelligence/Machine Learning Risk & Security Working Group
(AIRS), potential risks of AI can be categorized into data-related risks, AI/ML attacks,
risks related to testing and trust, and compliance (people) risks.

Data Related Risks

a. Learning Limitations

The effectiveness of any AI/ML system depends on the data used to train it and the
scenarios considered during the training. Sadly, training the system on all possible
scenarios and data is not always possible. For example, if we were to develop a model
that seeks to predict the occurrence of a major financial crisis, we would only have data
gathered from past crises. Even then, history shows that past financial crises were not
preceded by the exact same conditions or scenarios. There have always been unique
situations and “surprise” factors. In the end, we would develop a model that enjoys a
certain degree of accuracy, but it would be far from perfect.

Therefore, learning limitation is a key issue that’s often discussed during risk reviews
and brainstorming sessions that precede the deployment of an AI system.

b. Data Quality

Poor data quality limits the learning capacity of AI/ML systems leading to inaccurate or
unreliable output. Furthermore, poor data quality also negatively impacts future
decisions and inferences. Poor data may lead to inaccurate predictions and even failure
to achieve the intended objectives.

Potential AI/ML Attacks

Potential attacks against AI/ML systems include the following:

a. Data Privacy Attacks

If an attacker is able to infer the data used to train the AI/ML system, they may gain
access to sensitive data used to train the model, thereby compromising the privacy of
the system as a whole. We have two major types of data privacy attacks: membership
inference and model inversion attacks.

In a successful membership inference attack, an attacker is able to determine whether a

particular record is present in the training data set. They are able to determine, with a
certain degree of confidence, whether a certain input or set of inputs was part of the
data used to train the system. On the other hand, in a model inversion attack, an
attacker is able to extract representations of training data. They usually achieve this by
reconstructing training data from model parameters.

b. Training Data Poisoning

Data poisoning refers to the contamination of the data used to train the AI/ML system.
Data poisoning may increase the error rate and thus negatively impact the learning
process and the overall output. “Label-flipping’’ and “frog-boil’’ attacks are examples of
attacks under this category.

c. Adversarial Inputs

Some AI systems use inputs from external systems or users. These AI systems
interpret such input and undertake some actions, such as classifying the data. An
adversary could potentially deploy malicious algorithms designed to bypass the AI
system classifier. Most adversarial attacks usually aim to deteriorate the performance of
classifiers on specific tasks by essentially “fooling” the machine learning algorithm. Such
malicious inputs are referred to as adversarial inputs.

d. Model Extraction

In a model extraction attack, a malicious user attempts to steal the model itself. The
adversary first accesses the prediction API of the target model and then queries to
extract information about the model’s vital components. The goal is to use this
information to gradually train a pseudo model that works pretty much like the target
model. Attempts can then be made to sell the pseudo model. In some cases, the
adversary may take time to study the model even more closely with the aim of launching
further attacks on the parent AI system.

Testing and Trust Risk

One of the trickiest aspects of AI/ML systems is that some issues do not come to light in
the early stages of implementation. Rather, these issues become apparent after
continued use. The AI/ML system might also evolve and generate complexities that
might worsen over time. Here are potential concerns related to testing and trust risk:

a. Incorrect Output

Certain AI/ML systems are inherently dynamic and prone to changes over time. In
particular, the output may evolve over time and differ significantly from the output
produced in the early stages of implementation. This poses a real challenge to the
testing and validation of the AI/ML system. It may not be possible to carry out reliable
tests for all the scenarios, combinations, and permutations of the available data.

b. Lack of Transparency

Although AI/ML systems have been with us for a while, they have still been considered
an emerging technology with which most people are not yet fully conversant. Some
people have a very different understanding of how these systems work, a situation that
has led to persistent trust issues. It is believed, for example, that AI systems work out
problems in a ‘’black box’’ that isn’t open to scrutiny.

c. Bias

AI bias refers to injustice or unfairness against a person, entity, or corporation. Evidence

shows that AI systems usually churn out biased outcomes that can have serious
negative effects on individuals and organizations. Technologies are not neutral; they are
only as good (or bad) as the people who develop them. For example, in 2014, software
engineers at Amazon developed an employee recruitment program that used the
applicants’ resumes as input. However, the program was found to be biased against
women for technical roles. This discrimination forced the company to retire the program
prematurely.

Biased AI outcomes can also lead to legal, regulatory, reputational, and operational
risks.

Compliance
Policy Non-compliance

As the implementation of AI continues apace in the financial industry, there is a need to

consider its effects on the existing internal policies. Indeed, regulatory authorities have
expressed a lot of interest in AI, and multiple working groups have been formed to
discuss supervisory challenges posed by emerging technologies. This has led to a trove
of guidelines, white papers, and surveys on the subject as regulators seek to lay bare all
the emerging challenges and how they impact their work.
The Four Core Components of AI Governance
As the use of AI becomes more widespread within the financial industry, there’s a
general consensus among regulatory bodies and individual firms that there’s a need to
keep a close eye on AI emerging risks. However, there’s also an acknowledgment that
there are multiple ways to govern these risks. Each firm should be allowed to develop or
modify its own risk management framework to recognize these risks. There are four key
components of AI governance: definitions, inventory, policy/standards, and framework,
including controls.

Definitions
Definitions of AI/ML may vary from one organization to another depending on the
organization’s culture, environment, and adoption. The first step toward sound and
robust AI governance should include a clear definition of what constitutes AI (and what
doesn’t). This definition is vital since it provides the foundation and a clear
understanding of the other AI governance components.

For the best results, AI definitions should include the following:

● All the techniques an organization uses to train and develop AI.

● The features that distinguish AI from traditional rule-based systems.
● The implications of each definition.

The definitions and the supporting documentation should also clearly indicate how an
organization’s stakeholders identify with the AI definitions. Such stakeholders include
senior management, system developers, and legal, compliance, and information
security officers.

Inventory
An AI inventory is simply a centralized repository that helps an organization keep track
of all AI systems in use and monitor associated risks. An inventory describes the role of
each AI system deployed, its uses, and any restrictions on such use. Inventories can
also provide a list of data elements that constitute each AI/ML system.

Policies
In some cases, the use of AI systems can be governed on the basis of existing policies
and standards. However, there may be a need for the formulation of new policies and
standards or some modification of the existing ones to ensure that AI is deployed
appropriately.

It’s difficult to discuss AI policies and standards without mentioning ethical principles
and accepted norms. It should be noted that ethical principles for AI have been
discussed in the financial industry for a long time. As a result of these discussions, the
financial industry has developed a binding set of principles. Indeed, some institutions
have gone as far as publicizing the principles by which they abide. These principles
have had positive impacts on organizations and should therefore be developed further.

Framework
A robust AI governance framework is important as it helps organizations learn, govern,
monitor, and develop the AI system. The first part of an AI governance framework might
involve the identification of key stakeholders, formalized in the Center of Excellence
(CoE), working group, or council. The stakeholders for various groups and departments
collectively form what we call a ‘coalition.’ With such groups, best practices, sharing of
knowledge, and guidance on the use of AI systems can be achieved. Such efforts, in
most cases, bear fruits when close links are established with technology, data
engineers, and business line stakeholders.

In reviewing an AI-enabled initiative, the ‘coalition’ should consider the following:

● Data ethics.
● Privacy rights.
● Suitability of the training data used.
● Safety of the dataset.
● Supervisory roles.
● Data development (whether internal or external).

It should be noted that identifying the potential AI/ML risks helps formulate an
operational risk and control framework. Once potential risks have been identified, a gap
analysis can be established against existing controls. The number of participating
control owners depends on an institution’s control library. Thorough planning and a
structured approach are necessary to achieve the gap analysis. The gap analysis
results are then used to create new or improved controls to mitigate the identified AI/ML
risks.

Other factors to be considered under the AI framework include:

Monitoring and Oversight

A central monitoring process is important since it provides exposure to the decisions
made and the opportunity to raise concerns or challenges appropriately. It is, therefore,
necessary that the structure considers the changing needs of an organization as the
adoption of AI matures or as changes occur in the industry. Some firms believe that
monitoring and oversight procedures already in place sufficiently address potential AI
risks.

In most cases, existing governance is designed for scenarios where the degree of
human involvement is high. The accuracy, consistency, and efficiency of the existing
processes may be improved by reducing or removing interventions.

Third-party Risk Management

When deploying an AI/ML system, an organization may involve a dedicated third party
who has the knowledge and expertise required to pull off a successful launch. Such a
move also enables scalability, increased computing power, and increased access to
vendors within the larger fintech system. It is, therefore, necessary that firms strengthen
the capabilities of their third-party risk management (TPRM). In some cases, institutions
may include third-party contractual clauses with respect to the testing methodology of
the AI system.

Three Lines of Defense

A three-line-of-defense model includes:

● The first line: Responsible for business risk.

● The second line: Responsible for risk oversight and independent challenge.
● The third line: Tasked with independent assurance through the internal audit
function.
Roles and Responsibilities
An AI framework should clearly define the roles and responsibilities of various parties
within the organization. The following are examples of roles and responsibilities usually
considered within an AI framework:

Interpretability and Discrimination

Interpretability refers to the presentation of the AI system results in a format that a
human can understand. On the other hand, discrimination refers to an unfairly biased
outcome. The two form a major part of the risk management framework for any
organization deploying an AI/ML system. Let’s see the potential risks associated with
each.

Discrimination in AI
AI may lead to a discriminatory and unfairly biased outcome if not implemented
appropriately. Sources of such poor implementation include biased data, poor AI system
training, or the use of alternative AI systems or data sources that could potentially be
used to generate better outcomes for certain disadvantaged groups. An AI system that
may cause unfairly biased outcomes is likely to cause regulatory non-compliance issues
and legal and reputational risk.
Existing Legal and Regulatory Framework
Discrimination in areas that affect our day-to-day life, e.g., housing, lending, etc., is
prohibited by both federal and state statutes.

According to federal banking regulators, discrimination can be of three types: overt

discrimination, disparate treatment, and disparate impact.

Overt discrimination occurs when an AI system openly or actively discriminates. For

example, a lending model might be designed to give people from certain regions instant
“free” points even before the rest of their data has been considered. The key issue here
is that the model is obviously or blatantly providing or offering more favorable terms to
one group at the expense of another. However, overt discrimination doesn’t have to be
intentional. For example, a loan department might come up with a product that can only
be accessed by applicants above a certain age but, in so doing, lock out younger
applicants who may well have attained the legally accepted age.

Disparate treatment discrimination is the treatment of members of a protected class

differently from members of an unprotected class. In the context of AI lending, it occurs
when a model treats an applicant unfairly compared to other applicants based on the
applicant’s personal characteristics, e.g., race, gender, or sexual orientation. In an AI
system, a good example would be when an insurance company uses an AI model that
favors white over black people when assessing eligibility for coverage. It should be
noted that a system could be statistically sound but not legally non-discriminatory.

Disparate impact discrimination occurs when an AI system uses a neutral factor to

make a decision that affects a protected class more than an unprotected class. For
example, a lending model might give a loan to an applicant from a majority-white zip
code while turning down a similarly situated applicant from a majority-black zip code.
The neutral factor here is the zip code.

Data as a Cause of Discriminatory AI

Input data may lead to AI-related illegal discrimination in several ways:

● If it identifies or closely proxies class membership.

● If it causes protected class members to experience less favorable outcomes.
 ● If it has features that skew predictions for the protected class.

Traditional data inputs, for example, credit bureau attributes, have a lower probability of
causing disparate impact. This is because they are thoroughly vetted prior to approval
and publication for use by lenders. On the other hand, non-traditional data, for example,
rental payments, may raise more disparate impact concerns as compared to traditional
data. Such data usually raises coverage and accuracy concerns.

Algorithms as a Cause of Discriminatory AI

The complexity and opacity of algorithms may lead to discriminatory outcomes.
Algorithms may create interactions between variables and non-linear relationships that
are too complex for humans to understand. Such relationships may cause disparate
treatment by creating proxies for protected class status. However, some of these
concerns have been addressed by the use of AI methods that allow for a
comprehensive understanding of such complex relationships.

System misspecification may also lead to discriminatory outcomes. In this case,

prediction features for both outcome and protected class status may be independent,
but the class effect is included in the prediction.

As an example, assume that a lending model takes a loan applicant’s shopping habits
into account, particularly whether they buy goods at a discount store. At first thought,
this might look like an objective variable because it can reasonably be viewed as a
measure of wealth and, therefore, a predictor of repayment. But if the system goes
ahead to capture the store’s location, it may unintentionally capture a race effect
because different neighborhoods have different racial makeups. In this scenario,
shopping as a variable may serve as a proxy for the neighborhood, which in turn acts as
a proxy for race.

Interpretability/Explainability
Inconsistent Explanations
Unlike traditional linear systems, the same training data may be used for training
multiple AI/ML systems. Although the output from such systems is likely to be similar,
each model is likely to have a unique logical explanation as to how the AI output was
generated. The presence of different logical explanations for the same outcome can
ignite debate and serious discussions among the system’s developers and monitoring
teams.

Interpretability methods enhance human understanding of the AI/ML system, which

helps mitigate risks associated with the use of AI/ML systems.

Detection and Appeal of Incorrect Decisions

At the core of most AI/ML systems lies probabilistic tools that help the system make
decisions based on the likelihood of each set of events. But this means the system may
make incorrect decisions because the probability is not predictable. Even if the system
shows there’s a 50% chance a borrower will default, there’s simply no way to accurately
predict the outcome. The variables considered when computing the probabilities also
play a big role in all this. The more unrealistic such variables are, the more likely it is to
end up with an incorrect decision.

In some more complicated systems, neither the developer nor the user has a clear
understanding of the decision made or whether it is right or wrong. Thus, the
interpretability of high-impact AI/ML decisions is a huge source of risk. If there are
doubts over the correctness of an AI-driven decision that has a major impact on
individuals or the organization as a whole, there will be attempts to improve the model
or even replace it entirely to mitigate the effects.

Security Audit
Malicious actors could potentially misuse AI/ML. Interpretability is vital in ensuring that
AI/ML systems are protected as security evolves in the AI/ML world. The red team or
white-hat hacking audits in testing AI/ML systems may apply post hoc explanation
techniques in attacks against AI/ML systems.

Regulatory Compliance
Several legal regulations may require the use of interpretable systems, post hoc
explanations, and the documentation they facilitate. Such legal regulations include the
Equal Credit Opportunity Act, the Fair Credit Reporting Act, and the EU. General Data
Privacy Regulation and so on.

Common Practices to Mitigate AI Risk

To mitigate AI risk, there are three main areas of interest: the training data, the learning
procedure, and the output predictions. This gives rise to three corresponding risk
management approaches: pre-processing, in-processing, and post-processing
approaches. Post-processing approaches are suitable for runtime environments since
they do not necessarily require access to the training data. Furthermore,
post-processing approaches operate in a black-box environment, meaning that they
don’t need access to the internals of models. Thus, they can be applied to any machine
learning model.

Oversight and Monitoring

Oversight Processes
Oversight of an AI system capped by intensive monitoring to validate various aspects of
the system helps ensure the accuracy and efficiency of the system. An oversight
process can begin with the creation of an inventory of all the AI systems at a given
institution, the uses of the system, techniques employed, developers’ names, and risk
ratings. The evaluation includes assessing the inputs, outputs, and the AI system itself.
Assessing training data is important for ensuring data quality as well as identifying
potential biases that the data may contain. To evaluate the AI system, it is benchmarked
against optional models and known methods are utilized to ensure the interpretability of
the model.

Monitoring for Drift

Drift may result in a number of errors and risks in AI systems. Detection of drift can help
mitigate some AI-based risks. Monitoring helps to provide insight into the ‘‘accuracy
drift’’ of the model by estimating the accuracy of the model. Monitoring also helps to
provide insight into the “data drift” by checking for the deviation of the input data from
the training data.

Accuracy drift could worsen your model, while data drift, on the other hand, helps
businesses to understand the change in the characteristics of the data at runtime.

Addressing Discrimination in AI
The review of input variables and systems for evidence of discrimination in most lending
organizations is done by compliance, fair lending, and system governance teams.
Technological advances can enable the automation of most of these tasks.
Nevertheless, a human-centric approach may be required for a fair AI. It is not possible
for an automated process to fully substitute the experience and knowledge of a
well-informed team reviewing the AI system for discrimination bias. Therefore, it is
important that the first line of defense against discrimination in AI should include some
manual review.

A number of recently researched algorithms have proven to reduce class-control

discrepancies while maintaining the predictive quality of the system. In order to reduce
these discrepancies, the mitigation algorithms find the “optimal” system for a
corresponding quality and measure of discrimination.

Variables that cause discrepancies are excluded from the systems. Other tested
variables are used in their places. However, these methods have been shown to have
low rates of success in complex AI/ML systems.

More recently developed approaches for minimizing discrimination involve engaging in

data processing, making decisions within the algorithm, and carrying out
post-processing on the output.

Enhancing Interpretability and Explainability

Ensuring Quality Explanations
One of the major challenges to many institutions is ensuring that the explanations of
AI/ML are reliable and useful. Rough estimates and inaccurate or inconsistent
explanations in financial service institutions, for example, raise special concerns,
especially for credit lending decisions.

To reduce explainability-based risks, institutions may test the explanatory techniques

used for accuracy and stability on the simulated data.

Potential Risk Mitigation

Malicious actors could use available AI system explanations and predictions to attack
an organization. Organizations can mitigate such potential risks by only sharing the
required information. For instance, an institution may only share information that a given
consumer requires. Similarly, a firm should strictly share information whose
dissemination is legally required.

Strong traditional technology and cyber control could be used for effective AI-based risk
mitigation. The use of strong information security practices and watermarking could help
mitigate model extraction attacks. Watermarking involves training the AI/ML system to
produce unique output for a given input.

Practice Question
A financial firm has deployed a sophisticated AI system for credit scoring. Recently, the
firm’s IT department detected anomalous system behaviors. The anomaly report
showed that certain borrowers’ credit scores had been unusually boosted during the
model’s retraining phase. On further investigation, it was found that the scores were
manipulated through the subtle addition of incorrect labels to the training data. The IT
department suspects an orchestrated attack to distort the system’s learning process.
Which of the following potential AI/ML attacks is the firm most likely experiencing?

A. Membership inference attack

B. Model inversion attack

C. Training data poisoning attack

D. Adversarial inputs attack

The correct answer is C.

Training data poisoning attack typically involves the contamination of the AI/ML
system’s training data in a way that negatively influences its learning process or output.
In the given scenario, incorrect labels were subtly added to the training data during the
model’s retraining phase. This is a clear indication of a Training Data Poisoning Attack
aimed at manipulating the system’s learning process and consequently altering the
borrowers’ credit scores.

Option A is incorrect because a membership inference attack focuses on an attacker’s

attempt to ascertain whether a specific record was included in the training data set used
for the AI system. In the scenario provided, there’s no indication that such an inference
is being made; the problem revolves around the manipulation of training data, not the
disclosure of whether certain records were included in the training set.

Option B is incorrect because a model inversion attack involves an attacker extracting

specific information about the training data directly from the model. While this type of
attack also concerns the training data, the scenario described does not involve an
extraction of data from the model, but rather a manipulation of the labels within the
training data to distort the AI system’s learning process.

Option D is incorrect because an adversarial inputs attack typically involves the

intentional provision of malicious inputs designed to bypass the AI system’s
classification or decision-making mechanisms. In the presented scenario, the problem is
not related to the input data being used by the AI system after its training but rather
pertains to the alteration of the training data itself.
The Future Monetary
System
After completing this reading, you should be able to:

● Identify and describe the benefits and limitations of crypto and decentralized
finance (DeFi) innovations.
● Describe the role of stablecoins in DeFi ecosystems and differentiate among the
types of stablecoins.
● Discuss possible advantages and disadvantages of a monetary system based on
CBDCs.
● Understand the risks posed by the centralization that occurs in DeFi ecosystems
and crypto exchanges (CEX).
● Outline the regulatory actions recommended by the BIS to manage risks in the
crypto monetary system.

The monetary system is used by people all over the world to conduct payments and
other financial transactions. The organizations and structures that surround and support
monetary exchange make up the monetary system.

The central bank, which maintains the system’s fundamental functions and issues
money, is at the center of the monetary system. Trust in the central bank ultimately
underpins confidence in the monetary system.

The majority of payments are carried out by commercial banks and other private
payment service providers (PSPs), who also provide services to customers. This
separation of roles encourages competition.
Although the central bank-centered monetary system has done a good job historically,
new demands are being placed on it by technical advancements brought on by digital
innovations such as those in the crypto universe.

Decentralization is a foundational principle in the crypto universe. To make the system

self-sustaining and free from the influence of strong entities or groups, crypto envisions
checks and balances provided by several anonymous validators. Decentralized finance,
also known as “DeFi,” aims to replicate traditional financial services in the crypto
universe.

Recent occurrences have shown that the crypto ideal and its actuality are different.

● The collapse of the TerraUSD stablecoin and its companion coin Luna has
highlighted the system’s fragility.
● The crypto world, in particular, lacks a nominal anchor. It has become apparent
that crypto has deeper structural limits that prevent them from achieving the
effectiveness, stability, or integrity levels necessary for a suitable monetary
system.
● It is prone to fragmentation, and scaling its applications without sacrificing
security is impossible, as shown by the high fees.
● The system’s use of unregulated intermediaries is of grave concern.

Retail, fast payment systems (FPS) and central bank digital currencies (CBDCs), and
decentralization and permissioned distributed ledger technology (DLT) are
well-positioned to serve the public interest. They do so by lowering costs and increasing
convenience while upholding the integrity of the system. Faster, safer, and less
expensive payments and settlements are supported by these developments
domestically and internationally.

What do we want from a monetary system?

The monetary system, which comprises money and payment systems, is a collection of
institutions and structures that facilitate monetary exchange.
 ● Safety and stability: Money must perform the following three functions in order to
guarantee the system’s security and stability: a store of value, a unit of account,
and a medium of exchange.
● Accountable: Key entities that are dependent on the financial system must be
held accountable, either through particular mandates for public authorities or
through appropriate regulation and oversight for private businesses.
● Efficient: To facilitate economic transactions on a large scale and at a cheap cost,
the monetary system should be effective and enable dependable, quick
payments.
● Financial inclusion: To spread the advantages of economic activity and
encourage financial inclusion, everyone should have access to affordable, basic
payment services, particularly transaction accounts.
● User control: The system must guarantee user control over financial data and
defend privacy as a fundamental right.
● Integrity: The system’s integrity must be protected by preventing illegal activities
such as fraud, financing terrorism, and money laundering.
● Adaptable: The monetary system must be aware of technological advancements,
anticipate user needs and future advances, and adapt to shifting demands from
consumers and enterprises. Additionally, it ought to promote innovation and
competitiveness.
● Open: The monetary system also needs to be open, interoperable, and
adaptable both domestically and internationally in order to better serve a world
that is becoming more connected.

Challenges faced by today’s monetary system include:

● Users’ changing needs and concurrent technological changes have highlighted

areas for improvement.
● Current payment methods can occasionally be complicated and expensive, partly
due to a lack of competition.
● Many individuals still lack access to digital payment methods, particularly in
developing and emerging market economies.
 ● Demands for improvements to institutional structures and technological
advancements to meet the system’s changing societal demands.

The Promise and Pitfalls of Crypto

Stablecoins, which aim to tie their value to fiat currencies, highlight the crypto industry’s
desire to capitalize on the legitimacy offered by the unit of account issued by the central
bank. Stablecoins’ dependence on the legitimacy of central bank money is a major
structural weakness that is readily apparent. Additionally, stablecoins are less stable
than their issuers claim they are.

The abundance of coins also highlights another crucial weakness in the structure of
cryptocurrencies, namely their fragmentation into numerous incompatible settlement
levels.

The fragmentation of the cryptocurrency market reveals its challenge. The more users
flock to a single blockchain system, the worse the congestion and the higher the
transaction fees. Consequently, this invites newer competitors who might forgo security
in favor of greater capacity.

Crypto gives a sneak preview of potential functionality. These result from the ability to
bundle transactions and carry out the automatic settlement of bundled transactions in a
conditional manner, enabling higher functionality and speed.

The Building Blocks of Crypto

Bitcoin presented a revolutionary concept that gave rise to cryptocurrency: a
decentralized method of transferring value on a permissionless blockchain. Any
participant acting as a validating node can validate transactions on a public ledger.
Record-keeping on the blockchain is carried out by various anonymous, self-interested
validators as opposed to depending on trusted intermediaries (such as banks).
Cryptocurrency transactions are validated by decentralized validators and recorded on
the public ledger.

How does Crypto Work?

A buyer broadcasts the specifics of the transaction when a seller wishes to send
cryptocurrency to them. Validators compete for the opportunity to verify the transaction,
and whoever is chosen to do so adds the transaction to the blockchain. The exact
names of the parties involved in transactions are therefore kept a secret, but the history
of every transaction is publicly viewable and linked to particular wallets.

In order to make the system self-sustaining, miners and validators receive financial
rewards for carrying out their duties in accordance with the rules. Cryptocurrency
rewards can be received in the form of transaction fees or from rentals earned by
“staking” one’s coins in a blockchain that uses proof-of-stake. The more frequently a
node acts as a validator, and the higher the rents are, the higher the stake.

Many blockchains and cryptocurrency coins have emerged since the creation of Bitcoin,
most notably Ethereum, which enables the use of “smart contracts” and
“programmability.” The use of self-executing code known as “smart contracts” can
automate market processes and do away with the middlemen who were previously
needed to make choices. Smart contracts are transparent and less susceptible to
manipulation because the underlying code is open to the public and can be examined.
The ability of smart contracts to integrate many parts of a system is a key aspect. Users
can conduct complicated transactions on the same blockchain by merging several
instructions into a single, smart contract. By “tokenizing,” they can turn assets into
digital representations.

Since smart contracts cannot directly access data that is “off-chain” or outside the
specific blockchain, they depend on intermediaries to supply this information (so-called
oracles). But these modifications also bring new issues. More centralized validation
techniques are frequently used to boost capacity, which weakens security and
concentrates insiders’ benefits.

Stablecoins in Search of a Nominal Anchor

With the stated goal of reshaping the financial system by eliminating middlemen and
lowering costs, decentralized finance provides financial services and products. Lending,
trading, and insurance are the three major forms of financial activity, though the DeFi
ecosystem is expanding quickly.
Lending: Users can make interest-bearing loans of their stablecoins to borrowers who
pledge other cryptocurrencies as security on lending sites.

Trading: Decentralized exchanges (DEXs) are venues for direct trades between holders
of cryptocurrencies or stablecoins.

Insurance: Users of DeFi insurance platforms can protect themselves against things like
improper handling of secret keys, exchange hacks, and unsuccessful smart contracts.

Due to their pegs to fiat money or other assets like gold, stablecoins are essential to the
DeFi ecosystem. They value the credibility that the asset or unit of account issued by
the central bank offers. Their primary use case is to combat the extreme price volatility
and limited liquidity of unbacked cryptocurrencies like Bitcoin. Their use also prevents
costly, frequent conversions between cryptocurrencies and bank deposits in fiat
currencies.

The two main types of stablecoin are asset-backed and algorithmic.

Asset-backed stablecoins like Binance USD and USD Coin are typically administered by
a central intermediary who invests the underlying collateral and organizes the
generation and redemption of the coins.

Algorithm stablecoins. To sustain their value in relation to the target currency or asset,
algorithmic stablecoins rely on sophisticated algorithms that automatically rebalance
supply.

The reliability of stablecoin stabilization methods heavily rests on the quality and
transparency of their reserve assets, both of which are frequently terribly deficient.

Stablecoins lack the characteristics required to support the upcoming monetary system.

● They cannot make use of the legal constraints and protections of bank deposits
or the central bank’s role as a lender of last resort.
● They must import their credibility from sovereign fiat currencies.
● They restrict liquidity and risk fracturing the monetary system.
Structural Limitations of Crypto
The inherent limits of permissionless blockchains, which eventually result in system
fragmentation, congestion, and excessive costs, are a problem for cryptocurrencies.
Transactions are recorded on the blockchain by self-interested validators.

In order to maintain the decentralized consensus system, validators must be motivated

by financial rewards that are sufficiently high. Sincere validation must generate more
rewards than possible gains from cheating. If compensation becomes insufficient,
validators may be tempted to cheat and compromise the security of the cryptocurrency.

The blockchain’s capacity must be restricted in order to maintain incentives for

validators and keep fees high enough. In times of congestion, users may offer larger
fees to have their transactions processed more quickly because validators have the
discretion to determine which transactions are approved and processed.

The so-called scalability trilemma is manifested in the restricted scale of blockchains.

Only two of the three properties—scalability, security, or decentralization—can be
attained by permissionless blockchains by virtue of their design. Decentralization and
incentives improve security, but maintaining incentives through fees causes congestion,
which restricts scalability. As more recent blockchains that compromise on security have
entered the crypto universe, the restricted scalability of blockchains has caused the
crypto world to become fractured.

There are increased dangers of theft and hacking in a system of rival blockchains that
are not interoperable but are supported by speculation. Interoperability refers to the
ability of users to access, share, and validate transactions across several blockchains.
Since each blockchain represents a distinct settlement record, interoperability is not
possible in practice. To enable the movement of coins between blockchains, a few
“cross-chain bridges” have emerged. These bridges depend on a small number of
validators who, in the absence of regulation, must be believed to refrain from engaging
in unethical activity. However, a surge in hacks has coincided with the growth of these
bridges.
In contrast to the network effects that develop in conventional payment networks, the
crypto world is highly fragmented. In a conventional payment network, a platform gains
more users as more individuals use it. Trust and affordability are essential to such
platforms. Contrarily, cryptocurrency’s propensity for fragmentation and excessive fees
is a fundamental structural fault that disqualifies it from serving as the basis for the next
generation of money.

High price correlations between various cryptocurrencies and blockchains might be

caused by speculation. Increased demand drives up prices even further. Strong price
co-movement is present. When the rate of new users suddenly declines, there are
significant worries about what would happen to a system that depends on selling new
coins.

The DeFi Decentralization Illusion and the Role of

Exchanges
The DeFi ecosystem exhibits a propensity toward centralization despite its name. Voting
is used to make a number of important decisions among the owners of “governance
tokens,” which are frequently given to developer teams and early investors and are thus
highly centralized.

Since it is impossible to specify in contracts what to do in every scenario, conflicts must

be settled by some central bodies.

Additionally, newer blockchains typically rely on concentrated validation processes to

increase throughput and speed up transactions.

Incentives conflicts and hacking danger arise as validators become more centralized.

Furthermore, there are currently no regulations governing the screening of Oracle

providers. As such, anyone in control of Oracle can undermine the system by reporting
inaccurate data.

The cryptocurrency market is likewise centralized, with investors favoring centralized

exchanges (CEXs) over decentralized ones (DEXs). Similar to traditional banking, CEXs
keep off-chain records of the orders that traders have posted. Due to their cheaper
expenses, CEXs also draw greater trading activity than DEXs. Since 2020, CEXs have
experienced significant expansion and attained quantities that qualify them as important
from the financial stability perspective.

Before adopting relevant regulatory regulations, it is necessary to assess the parallels

and divergences between the crypto market and traditional finance.

Regulatory Approaches to Crypto Risks

To address the immediate risks in the cryptocurrency monetary system, regulatory
action is required.

Authorities need to take regulatory arbitrage seriously. They should make sure that
crypto and DeFi operations adhere to the legal requirements for similar traditional
activities, operating under the guiding premise of “same activity, same risk, same
regulations.” The recent collapse of numerous stablecoins has brought attention to how
urgent the situation is.

Policies are required to promote the integrity and safety of the financial and monetary
systems. A fine or closure should be imposed on cryptocurrency exchanges that
conceal the identities of parties to transactions and disregard fundamental
know-your-customer (KYC) and other Financial Action Task Force (FATF) rules. They
could otherwise be used to dodge taxes, finance terrorists, or get around economic
restrictions. Similarly, banks and credit card firms need to demand user identity and
carry out KYC compliance.

Consumer protection laws. Investors ought to be able to invest in riskier securities, such
as cryptocurrencies, but there needs to be sufficient disclosure. This entails sound
regulation of the frequently deceptive advertising of digital assets. Front-running-like
behaviors can necessitate the application of innovative legal strategies.

The exposure of banks and nonbank financial intermediaries to the cryptocurrency

ecosystem poses dangers to the stability of the financial system that central banks and
regulators must reduce. Traditional financial institutions are investing in cryptocurrencies
rapidly, meaning that shocks to the cryptosystem might have a ripple effect. Large
traditional banks have so far only had minor exposures, and their direct investments in
businesses engaged in cryptocurrency markets are still modest compared to their total
capital. Sound guidelines for bank exposures to cryptocurrencies must be implemented
to address these risks.

It is crucial to discover regulatory entry points and close data gaps. A multitude of new
centralized intermediaries has popped up as a result of the cryptocurrency market’s
expansion. The traditional financial institutions and these centralized entities serve as a
natural entry point for regulatory responses.

The worldwide nature of crypto will necessitate collaboration across all regulatory fields.
In order to sanction non-compliant actors and platforms, authorities may need to
actively exchange information and conduct coordinated enforcement measures.

Crypto’s Lessons for the Monetary System

Overall, the cryptocurrency industry offers a glimpse of exciting technological potential,
but it is unable to achieve all the major objectives of a digital monetary system. It has
flaws such as stability, effectiveness, accountability, and integrity that legislation can
only partially resolve. Fundamentally, cryptocurrency and stablecoins result in a
fragmented and weak monetary system that comes from the economics of incentives
rather than technological limitations.

The goal is to harness the beneficial cycle of network effects by not only enabling
valuable functionalities like programmability, composability, and tokenization but also
putting them on more secure bases. Central banks are utilizing the best aspects of new
technology to serve the public good, along with their attempts to regulate the crypto
universe and address its most pressing issues.

Vision for the Future Monetary System

New technical advancements and a more accurate representation of central bank
money at its core should combine to create the future monetary system. The benefits of
new digital technologies can therefore be gained through interoperability and network
effects because they are rooted in trust in the currency.
As one of its primary functions is to issue central bank money, which acts as the basis of
the future monetary system and the unit of account in the economy, central banks are
ideally positioned to provide it.

Using its balance sheet, the central bank’s other function is to provide the means for the
ultimate finality of payments. It’s a trusted middleman who debits the account of the
ultimate payer and credits the account of the ultimate payee.

Another function of the central bank is to ensure enough liquidity available for
settlement to support the smooth operation of the payment system. L liquidity provision
ensures no bottlenecks will obstruct the smooth operation of the payment system.

The central bank’s other responsibility is to protect the payment system’s integrity
through regulation, supervision, and oversight. The central bank is in a good position to
provide the groundwork for cutting-edge services in the private sector thanks to its
interconnected roles.

The future monetary system expands on these central bank functions and
responsibilities in order to provide room for new capabilities of central bank money and
cutting-edge services to be built on top of them. New private applications will be able to
function on more advanced technology representations of central bank money rather
than stablecoins. Thus, innovations at the central bank can support numerous new
activities.

Components of the Future Monetary System

The future monetary system will be based on the tried-and-true division of labor
between the central bank, which supplies the framework of the system, and the private
sector organizations, which handle customer-facing operations. Additionally, new
standards, like application programming interfaces (APIs), can be created to
dramatically improve the interoperability of services and related network effects.

This vision includes elements at the wholesale and retail levels that make a variety of
new features possible. At the wholesale level:
Programmability. It can facilitate transactions between financial intermediaries and
provide new capabilities. Using permissioned distributed ledger technology (DLT),
CBDC transactions allow programmability and settlement, enabling transactions to be
conducted automatically when predetermined conditions are met.

Composability. Make it easier for transactions to be composed by enabling the

combination and execution of a variety of different functions.

A wider range of financial intermediaries. New capabilities not only enable a far larger
range of financial intermediaries to participate in transactions but also allow for an
increase in the types of transactions.

Multi-CBDC arrangements. Additionally, Central Bank Digital Currencies (CBDCs)

collaborate internationally through multi-CBDC agreements, including various central
banks and currencies.

Tokenization. When deposits are tokenized, they are decentralized, settled, and given a
digital representation on the distributed ledger technology (DLT) platform. This might
make it easier to exchange new technologies, including fractional ownership of stocks
and other assets, opening the door to creative financial services that go far beyond
simple payment processing.

At the retail level:

Interoperability. Users of one platform can send messages and commands to others
that are linked to it with ease through APIs. Retail innovations encourage more
competition, decreased costs, and increased financial inclusion in this way.

Instant payments. Households and businesses can easily access retail CBDCs.
Through various interfaces and competing private Payment Service Providers, they
enable immediate payments between end users.

Data architecture. The Central Bank Digital Currencies (CBDCs) are backed by a data
architecture that includes digital identity and APIs that allow for safe data interchange,
enabling more user control over financial data.
Open platform. They facilitate lower costs in payment services by fostering efficiency
and better competition among private sector PSPs through the provision of an open
platform.

Inclusive design. Both can help financial inclusion for people who do not already have
access to digital payments through inclusive design elements.

A Metaphor for the Future Monetary System

The central bank serves as the solid base of the metaphorical tree that represents the
future monetary system. The tree metaphor illustrates that the monetary system is
founded on payment finality through eventual settlement on the central bank’s balance
sheet.

The central bank-based monetary system fosters a thriving ecosystem of participants

and activities where rival private-sector Payment Service Providers can employ
creativity and innovation to serve users better.

When viewed from a distance, the world’s monetary system can be compared to a
forest, whose canopy makes cross-border and cross-currency activities possible.
Infrastructures like multi-CBDC (Central Bank Digital Currencies) platforms function as
crucial new system components in the canopy.

Innovation is crucial. For private innovation to serve the public good, the fundamentals
must be done correctly. Public infrastructures can enhance the payment system by
utilizing many of the purported advantages of cryptocurrencies without the downsides,
as demonstrated by ongoing work at central banks. How central banks may assist
interoperability and data governance is demonstrated through wholesale and retail
CBDCs as well as open banking initiatives.

Wholesale CBDCs and Tokenized Money

A Central Bank Digital Currency (CBDC) is a digital payment instrument that is directly
owed to the central bank and is valued in the country’s unit of account. Retail CBDCs
are available to both individuals and companies. However, unlike domestic commercial
banks, wholesale CBDCs provide additional payment and settlement capabilities to a
considerably wider range of intermediaries, potentially spurring innovation.

Multi-CBDC systems involving several central banks and currencies can benefit from
decentralized governance. Functions such as self-executing smart contracts are
essential because they enable users to limit the settlement of their transactions to the
event that certain predefined conditions are met (programmable). Through such
automation, payments, and delivery of securities can be done only all at once or not at
all, accelerating settlement and reducing counterparty risk.

Wholesale CBDCs have the advantage of potentially being accessible to a far broader
range of intermediaries than domestic commercial banks. By allowing nonbank PSPs to
conduct CBDC transactions, competition, and vitality could increase significantly.

Payments could be finalized with wholesale CBDCs. In a digital system, establishing the
source or “provenance” of the transferred funds is necessary to ensure that payment is
genuine. Cryptography is used to conceal real names while openly displaying the
complete history of all transactions made by all parties.

The tokenization of regulated financial instruments, including retail deposits, can be

supported by central banks. Commercial bank deposits can be represented digitally
through tokenized deposits. Tokenized deposits would be programmable, “always on”
(24/7), and covered by deposit insurance, making them suitable for a wider range of
retail payment applications, such as autonomous ecosystems. This approach could
make it easier for other financial assets, like stocks or bonds, to be tokenized, enabling
fractional ownership of those assets.

One potential tokenized deposit system would include a permissioned platform that
keeps track of every token transaction the participating institutions produce. Retail
investors (depositors) would store tokens in digital wallets and send tokens between
wallets to make payments. This might occur concurrently as a result of smart
contract-enabled single atomic transactions. Digital representations of bonds and stocks
might potentially be possible with the same system.
In autonomous ecosystems, programmable CBDCs might also facilitate
machine-to-machine payments. Machines can manage their own budgets and make
direct purchases from one another for goods and services. Their integration will
enhance the need for programmable money and smart contracts, lowering any
settlement risk. For instance, this might result in huge efficiency savings in the goods
logistics industry, where transactions are still primarily paper-based and can take
several days.

The advantages of atomic settlement and open-source protocols can coexist peacefully,
with central banks serving as the main hubs for validation.

Retail CBDCs and Fast Payment Systems

Wholesale and retail CBDCs have a lot in common. Retail CBDCs provide consumers
and companies with digital access to money from the central bank. Retail-facing
payment services are provided by PSPs, both bank and nonbank. Thus, retail CBDCs
are occasionally referred to as “digital cash.” As transfers take place in real-time or
close to real-time, on an almost 24/7 basis, they have great speeds and availability.

Contrary to crypto, which needs high rents, has limited scalability, and is congested,
Central Bank Digital Cryptos and retail Fast Payment Systems (FPS) provide broader
use, which lowers costs and results in better services.

Retail Fast Payment Systems (FPS) have already made notable strides toward lowering
costs and promoting financial inclusion for the unbanked, particularly in Brazil, where
two-thirds of the adult population use their retail FPS Pix.

Retail CBDCs can be created to assist financial inclusion by examining retail CBDC
design elements that address particular barriers to financial inclusion, such as through
novel interfaces and offline payments. The cost of payment services for the unbanked
can be decreased by central banks through tiered CBDC wallets with simpler due
diligence for users transacting in lower values, promoting more access to digital
payments and financial services.

As every transaction in the digital economy leaves a trace, generating issues with
privacy, data misuse, and personal safety, retail Central Bank Digital Cryptos (CBDCs)
and Fast Payment Systems (FPSs) can be created with privacy protection and
improved user control over data in mind. Since there isn’t a single, comprehensive
record of every transaction in conventional payment systems, individual Payment
Service Providers (PSPs) have more control over data. Every PSP, however, only
records its own transactions. Even the central operator is unaware of all payments in
full. Thus, the promise of confidentiality by the central operator and the mix of
segregated record-keeping are used to protect the privacy of payments.

Crypto advocates claim permissionless blockchains give consumers back control over
their personal data, yet this poses serious threats to users’ privacy and integrity. Retail
FPSs and CBDCs’ underlying structures can allow users control over their personal
data while protecting their privacy and promoting consumer welfare. Additionally, central
banks can genuinely build systems in the public interest because they have no financial
interest in personal data.

Systems that provide law enforcement authorities with access to information while
maintaining the necessary legal protections can be created with clear mandates and
public accountability. These methods are currently being considered for retail CBDCs
and are already standard practice in the form of bank secrecy rules.

Retail CBDCs and FPS have the potential to enhance accountability in comparison to
the current system and the crypto universe. The role of the central bank in retail
payments needs to be discussed publicly as a result of new systems, and legal
regulations will need to be modified in order for them to operate. Private service
providers would be integrated into a strong regulatory and oversight framework under a
system based on public infrastructure.
Principles for the Effective
Management and
Supervision of
Climate-related Financial
Risks
After completing this reading, you should be able to:

● Describe the principles for the management of climate-related financial risks

related to corporate governance and internal control framework.
● Describe the principles for the management of climate-related financial risks
related to capital and liquidity adequacy and risk management process.
● Describe the principles for the management of climate-related financial risks
related to management monitoring and reporting, comprehensive management of
credit risk and other risks, and scenario analysis.
● Describe the principles for the supervision of climate-related financial risks
related to prudential regulatory and supervisory requirements for banks and the
responsibilities, powers, and functions of supervisors.

Financial risks associated with climate change can affect banks regardless of their size.
Therefore, banks should think about how climate-related risks affect their operations
and evaluate how financially significant these risks are. They must manage the financial
risks associated with climate change in a way that is appropriate for the scope and
complexity of their operations and the level of risk that each bank is willing to take.
Risks associated with the climate can have extensive effects. The distinct
characteristics of these risks, such as potential transmission channels, the intricacy of
the impact on the economy and financial sector, and uncertainty related to climate
change, should be considered.

The effects of climate change might manifest throughout a wide range of time and are
expected to get worse with time. A bank’s typical two- to three-year capital planning
horizon may not be long enough to account for all climate-related risks. Given how
unpredictable the timing of these risks is, banks should approach building their risk
management capabilities with caution and flexibility.

Banks should continually expand their knowledge and skills about financial risks
associated with climate change in line with the risks they confront. Further, banks
should make sure they have the resources set aside to manage these risks.

There is no recommended board structure in the Basel framework principles.

Corporate Governance
Principle 1: Banks should create and implement a robust approach for comprehending
and evaluating the potential effects of climate-related risk drivers. Banks should
consider significant financial risks associated with climate change that could arise
across a range of time frames and incorporate these risks into their overall business
plans.

When creating and implementing their business plans, banks should take material
physical and transition risk factors into account. This includes understanding and
assessing how these risks may affect how resilient a bank’s business model is over the
short, medium, and long terms. The board and senior management ought to be involved
in pertinent phases of the procedure. Besides, the managers and staff at the bank ought
to be made aware of the board’s strategy.

A bank’s strategy and risk management frameworks should consider the serious
financial risks associated with climate change. Besides, the board and senior
management should decide whether changing remuneration practices is necessary.
Banks’ risk management plans should be in line with their declared goals and
objectives. The board and senior management should make sure that their internal
strategy and risk appetite declarations are in line with any disclosed climate-related
strategies and commitments.

Principle 2: In addition to exercising effective oversight over financial risks associated

with climate change, the board and senior management should clearly define members’
and committees’ roles in managing climate-related risks.

Board members or committees should specifically be charged with managing

climate-related financial risks. This responsibility should be adequately considered as
part of a bank’s business strategy.

Banks need to make sure that the board and senior management are aware of the
financial risks associated with climate change and have the knowledge and experience
necessary to handle those risks.

Banks should clearly define and assign duties and responsibilities related to recognizing
and managing climate-related financial risks in their organizational structure and make
sure relevant business units have enough resources to carry out these obligations.

Principle 3: Banks should develop the proper policies and procedures and implement
them across the entire organization to achieve successful management of
climate-related financial risks.

All relevant activities and business units should implement policies, methods, and
controls that include management of major climate-related financial risks.

Internal Control Framework

Principle 4: Banks should include climate-related financial risks in their internal control
procedures to ensure solid, thorough, and efficient identification, measurement, and
mitigation of material climate-related financial risks.

A clear description and assignment of climate-related responsibilities and reporting

should be part of the internal control structure.
Climate-related risk assessments may be carried out throughout client onboarding,
credit application, and credit review stages. In addition, these assessments should run
during ongoing client interaction and monitoring as well as in new products or business
approval processes.

Independent of climate-related risk assessments, the risk function should be in charge

of conducting climate-related risk assessments and monitoring. This involves
questioning the initial evaluation made and ensuring that all applicable laws and
regulations are followed.

The quality of underlying data, the risk governance framework, the business and risk
profile, and the overall internal control structures and systems should all be
independently reviewed and objectively assured by the internal audit function.

Capital and Liquidity Adequacy

Principle 5: Banks should identify and quantify climate-related financial risks and
incorporate those determined to be material over relevant time frames into their internal
capital and liquidity adequacy assessment processes.

Banks should develop procedures to assess how climate-related financial risks that
could materialize within their capital planning horizons will affect their solvency.

It is equally important for banks to determine if climate-related financial risks could result
in net cash outflows or the depletion of liquidity buffers. Banks can do this by
considering extreme yet possible scenarios and including those risks in their internal
liquidity management strategies.

The incorporation of climate-related financial risks that have been deemed material also
entails the incorporation of physical and transition risks that are pertinent to a bank’s
business model, exposure profile, and business strategy. These ought to be evaluated
for inclusion in their stress testing programs as data over appropriate time horizons.

As the methodology and data used to analyze climate-related financial risks continue to
develop over time and analytical gaps are closed, it is believed that these risks will likely
be continuously incorporated into banks’ internal capital and liquidity adequacy
evaluations.

Risk Management Process

Principle 6: Banks should recognize, track, and manage all financial risks related to the
climate that could significantly deteriorate their financial position and capital resources
while making sure their risk management strategies consider all potential risks and
establish a solid plan for dealing with them.

The board and senior management should make sure that a bank’s risk appetite
structure clearly defines and addresses climate-related financial risks, where applicable.

To begin with, banks should routinely conduct thorough assessments of the financial
risks associated with climate change. They should establish such risks’ definitions and
thresholds for materiality, including the risks posed by concentrations in particular
sectors and geographical areas. Banks need to identify important risk indicators that are
appropriate for their regular monitoring and escalation procedures.

Further, banks should think about risk-reduction strategies. Among others, such
strategies could include setting internal caps for the many kinds of significant financial
risks related to climate change to which banks are exposed.

Lastly, banks should keep an eye on future developments. They should work to
understand and manage the impacts of climate-related risk drivers on other material
risks because there may yet be undiscovered channels for transmitting these risks to
traditional financial risk categories.

Management Monitoring and Reporting

Principle 7: For efficient board and senior management decision-making, banks should
work to ensure that their internal reporting systems are capable of monitoring significant
financial risks related to the climate and delivering timely information.

To make it easier to identify and report risk exposures, concentrations, and developing
concerns, a bank’s risk data aggregation capabilities should incorporate climate-related
financial risks. It should have mechanisms in place to guarantee the accuracy and
reliability of the gathered data. In addition, a bank should have systems in place to
gather and aggregate financial risk data connected to climate across the entire
company.

In order to better understand their transition strategies and risk profiles, banks should
think about actively engaging customers and counterparties and gathering more
information. In the absence of trustworthy or comparable information, banks may
consider adopting reasonable substitutes and assumptions.

Considering the dynamic nature of financial risks associated with climate change, banks
should determine an acceptable frequency for updating internal risk reporting.

To assess, track, and report financial risks associated with climate change, banks
should establish qualitative and quantitative measures and indicators. Any restrictions
that impede this should be made clear to the relevant stakeholders.

Comprehensive Management of Credit Risk

Principle 8: Banks should be aware of how climate-related risk factors affect their credit
risk profiles and make sure that their credit risk management systems and procedures
take these risks into account.

Banks should have carefully thought-out credit policies and procedures to address
significant climate-related credit risks. These include appropriate policies and
procedures to recognize, quantify, assess, track, report, and manage or lessen the
effects of significant climate-related risk drivers.

Banks should take a variety of risk-mitigation measures into account to control

significant climate-related credit risks. Such measures include altering credit
underwriting standards, using targeted customer interaction, or putting restrictions on
loans. Additionally, they need to put restrictions on the businesses, industries, or
geographic regions that they are exposed to that do not fit their risk tolerance.

Comprehensive Management of Market, Liquidity,

Operational, and Other Risks
Principle 9: Banks need to be aware of how climate-related risk factors affect their
market risk positions and make sure that market risk management systems and
procedures take important climate-related financial risks into account.

Banks should determine the risk factors related to climate change that are most likely to
have an impact on the value of the financial instruments in their portfolios. Besides, they
should assess the likelihood that losses will occur and the potential for increased
volatility and set up efficient procedures to limit or mitigate the resulting effects.

An analysis of a sudden shock scenario could be a helpful tool for better understanding
and evaluating the relevance of climate-related financial risks to a bank’s trading book.
Among other factors, such evaluation focuses on variation in liquidity across assets
exposed to climate-related risk and the speed at which exposures could reasonably be
closed out.

Banks may consider how the cost and accessibility of hedges could change when
assessing their mark-to-market exposure to climate-related risks.

Principle 10: The influence of climate-related risk drivers on banks’ liquidity risk profiles
should be understood, and banks should make sure that their systems and procedures
for managing liquidity risk consider significant financial risks associated with climate
change.

Banks should evaluate the effects of financial risks related to climate change on net
cash outflows or the value of the assets that make up their liquidity buffers. Banks
should take these effects into account when calibrating their liquidity buffers and when
developing their frameworks for managing liquidity risk.

Principle 11: Banks need to be aware of how climate-related risk factors affect their
operational risk and take the necessary steps to account for these risks if they are
significant. This comprises risk factors relating to the climate that could increase the risk
of strategic, reputational, and regulatory compliance.
When creating business continuity plans, banks should consider the material
climate-related risks that could have a significant impact on their operations and their
capacity to continue delivering essential services.

Banks should evaluate how climate-related risk drivers affect other risks, such as
strategic, reputational, regulatory compliance, and liability risks. Banks should take such
risks into consideration as part of their risk management and strategy-setting
operations.

Scenario Analysis
Principle 12: Banks should utilize scenario analysis to evaluate the adaptability of their
business models and strategies to a variety of scenarios and assess their impact on the
company’s overall risk profile. A variety of significant time horizons should be
considered when evaluating these.

The goal of climate scenario analysis should be in line with a bank’s overall goals for
managing climate risk. The scenario analysis may include examining how the bank’s
strategy and the shift to a low-carbon economy will be affected, as well as measuring
the bank’s vulnerability to these risks and calculating exposures and potential losses.

Scenario analysis should consider relevant climate-related financial risks. Besides, it

should encompass a variety of conceivable outcomes. Banks should also think about
the advantages and drawbacks of various scenarios and assumptions.

Banks should develop the capacity and knowledge necessary to carry out climate
scenario analyses that are proportional to their size, complexity, and business model.
Consequently, larger, more complicated banks should have stronger analytical
capabilities.

Scenario analysis should use a variety of time horizons, from short- to long-term. Risk
analysis can be conducted over shorter time periods with less uncertainty, while longer
time frames will have higher levels of uncertainty.
Climate scenario analysis is a very dynamic field, and the methods employed are
expected to change quickly. Models and findings from climate scenario analysis should
be challenged and reviewed frequently by a variety of internal or external specialists.
Machine Learning and AI
for Risk Management
After completing this reading, you should be able to:

● Explain the distinctions between the two broad categories of machine learning
and describe the techniques used within each category.
● Analyze and discuss the application of AI and machine learning techniques in the
following areas:
● Credit risk.
● Market risk.
● Operational risk.
● Regulatory compliance.
● Describe the role and potential benefits of AI and machine learning techniques in
risk management.
● Identify and describe the limitations and challenges of using AI and machine
learning techniques in risk management.

Machine learning is a branch of artificial intelligence (AI) that uses algorithms to identify
patterns in a data set and then make decisions, just like humans. It aims to imitate how
humans learn, gradually improving its predictive power and accuracy. Machine learning
is premised on the realization that machines can learn without being programmed to
perform specific tasks. Machine learning algorithms use statistical methods to uncover
key insights within a data set and then make relevant classifications or predictions.

In recent years, machine learning has gained a strong foothold in the financial industry,
particularly banking and insurance. It has been used to decide the amount of money to
lend to customers, provide warning signals to traders, detect fraud, and improve
compliance with rules and regulations. This chapter explores ways machine learning
and AI can improve risk management by leveraging the large volume of data available.
We also look at the core machine learning techniques which can be applied to improve
risk management.

Machine learning falls into two broad categories: supervised and unsupervised machine
learning.

Supervised Learning
Supervised learning is a machine learning technique where models are trained using
labeled data. The goal is to find the mapping function that maps the input variable (X)
with the output variable (Y).

Y = f(x)

The word “supervised” comes from the fact that the algorithms used aren’t left to
deduce the relationship between X and Y on their own. Instead, the machine is trained
using data that are already labeled. It’s pretty much like providing the machine with
some questions that are already tagged with the correct answers and then asking it to
find the answers to untagged but similar questions.

Categories of Supervised Learning

Regression Machine Learning
Regression machine learning is a technique that predicts a single output (dependent
variable) value using training data (independent variables). For example, we can use
regression to model the risk of loan repayment using a range of explanatory variables,
including average nonpayment rates, employment status, credit history, and other
outstanding liabilities.

One advantage of machine learning regression over traditional regression is that we can
include a larger number of independent variables that can be discarded automatically if
they lack any explanatory power. For example, LASSO regression eliminates variables
with zero regression power. In contrast, Ridge regression gives lower weights to
variables that are highly correlated with other variables in a model. We can also begin
with zero power for all variables and gradually add the variables found to have
explanatory power.

Supervised learning employs a technique known as Principal Component Analysis

(PCA) to simplify complex datasets. PCA is a statistical procedure that transforms
potentially correlated variables into a smaller number of uncorrelated variables called
principal components. The objective of PCA is to reduce the dimensionality of the data
while retaining as much information as possible.

Consider a scenario where you’re modeling credit repayment risk as the dependent
variable. Among the independent variables, you have (I) owns a house, (II) owns a car,
and (III) has bank savings. Each of these variables represents different aspects of a
person’s financial status and can potentially influence their credit repayment risk.

In a real-world dataset, there might be many such variables, making the analysis
complex and computationally intensive. This is where PCA comes in. Rather than
working with all these variables independently, PCA identifies the commonalities among
these variables and combines them into a smaller set of synthesized variables – the
‘principal components.’

In our example, PCA could combine the three original variables into a single variable –
asset ownership. This new variable captures the shared variance among the original
variables, thereby encapsulating the underlying construct of ‘financial stability’ that they
collectively represent. By doing so, PCA enables a more efficient and manageable
analysis without sacrificing too much valuable information.

It’s important to note, however, that the results of PCA must be interpreted cautiously.
The original meaning of the variables can get lost or distorted in the PCA process, and
the principal components might not always have a clear or intuitive meaning.

Classification
Classification involves grouping data into labeled classes. For example, when modeling
the likelihood of default, we could have two categories: Potential defaulters and
non-defaulters. The model would then be trained on how to classify the data into one of
the two classes in an accurate manner. In binary classification, the model works with
just two labels – 0 and 1 (yes and no). In the case of multi-class classification, the
model classifies data into more than one class.

Unsupervised Learning
In unsupervised learning, models are not supervised or trained using labeled data.
Instead, models find hidden patterns and insights from the given data without human
intervention. The goal is to identify previously undetected patterns and discover the
internal structure of a data set without predefined output categories. Unsupervised
learning methods are used to perform more complex processing tasks compared to
supervised learning. For example, a bank could create an algorithm to scrutinize
customer accounts and identify those with similarities. This could help the bank develop
a product that specifically targets those account holders.

Categories of Unsupervised Learning

Clustering
Clustering mainly deals with finding a natural structure or pattern in a collection of raw,
unclassified data. Unsupervised learning clustering algorithms scour the data to identify
any notable clusters (groups).

One area where clustering is applied is the detection of spam emails. If an email looks
like others deemed spam, it is also likely to be spam.
There are several clustering approaches, but the most popular one is k-means
clustering. Under k-means clustering, the desired number of clusters, k, is
predetermined. The algorithm is then tasked with clustering the data into the k groups
through an iterative process. A larger k means you’ve got smaller groupings with more
granularity, while a lower k means larger groupings with less granularity. Iteration is
aimed at maximizing the difference in means between determined groups. Each group
or cluster has its own centroid (central focal point). If we have two clusters, A and B,
and a data point Y is closer to the centroid (mean) of A than B, then Y is put in cluster A.

Dimensionality Reduction
Dimensionality reduction is used to analyze and obtain a better representation of data.
The data set should have less redundant information at the end of the process, but the
important parts may be emphasized. In practice, this technique is used to hive off a
section of a large amount of data for closer scrutiny.

In addition to supervised and unsupervised machine learning techniques, we have deep

learning and neural networks – other branches of machine learning that can be
supervised, unsupervised, or semi-supervised. The two are used to model super
complex relationships between variables and ultimately to better mimic human
decision-making. A key feature of deep learning is that problems are modeled in a
multi-layer network that is extremely difficult to comprehend. As the input data
progresses through the model, it’s combined and recombined to form new factors with
weights that depend on the combinations made in the previous layer. This leads to
output that’s essentially been worked out in a “black box.” This perceived lack of
transparency and clarity over decisions made by the model can complicate risk
management and can be a source of risk for firms. This is an issue that has widely been
mentioned in the digital lending market, where the software used essentially runs
borrower data through a black box to determine whether they are eligible for loans and
how much they should get.

The Application of AI and Machine Learning Techniques

Credit Risk
For a long time, firms relied on classical linear, logit, and probit regressions to model
credit risk. However, in recent years, there’s been a realization that AI and machine
learning can significantly improve credit risk management and help firms make better
lending decisions. Studies have shown that credit risk can be modeled more accurately
by combining traditional statistical methods of distress and bankruptcy prediction and
neural network algorithms.

One area where machine learning has proved extremely useful is the credit risk analysis
of credit default swaps. This is because, in the CDS market, there are many uncertain
elements involved in the determination of the likelihood of a default (credit) event and
the estimation of the cost of default in case a default materializes. In a 14-year study
conducted between 2001 and 2014 involving CDSs of different maturities and rating
groups, nonparametric machine learning models outperformed traditional benchmark
models regarding prediction accuracy. The nonparametric machine learning models
also performed better in terms of suggesting the most practical hedging tools.

Banks are increasingly relying on machine learning to make better consumer and SME
lending decisions.

Market Risk
Market risk emanates from exposure to the financial market, including investing and
trading in various assets such as stocks and bonds. Machine learning has been used in
several market risk management areas:

● Stress testing market risk models to discover hidden or emerging risks: One of
the risks that come with the use of models to analyze market risk is that such
models can be incomplete, false, invalid, or even incorrectly specified in terms of
parameters. Machine learning can reveal whether the models used have any of
these issues. Machine learning techniques can also be used to scan for
unsuitable assets in trading models. For example, yields.io offers an AI-driven
algorithm that provides real-time model monitoring, model testing for deviations,
and model validation.
● Understanding the impact of a firm’s trading activities on market pricing: AI and
clustering techniques have been used by large trading firms to look for ways to
 reduce transaction costs associated with large trades that have the potential to
move the market price, particularly in illiquid markets.
● Providing real–time warnings to traders: A combination of neural networks and
decision trees can be used to provide traders with real-time warnings of
impending changes in underlying trading patterns.

Operational Risk
Operational risk concerns itself with risks emanating from both internal and external
operational breakdowns. Such risks are attributable to people (e.g., strikes and
go-slows), systems, frauds, neglected procedures, or natural disasters. In recent years,
operational risks have become more complex and frequent, prompting firms to explore a
path toward artificial intelligence and machine learning-based solutions.

AI and machine learning can help firms to:

● Identify, measure, estimate, and assess the impact of operational risk exposures.
● Identify effective risk mitigation strategies.
● Find instruments that can facilitate shifting or trading risk.

AI and machine learning can be effective tools against fraud. This happens when firms
automate routine tasks to minimize human error. Besides, machine learning and AI can
be employed in processing unstructured data to screen out relevant content or negative
news and evaluate the extent of interconnectedness among individuals to assess how
prone they might be to an external attack. Further, AI tools can be used to monitor
individual traders by combining trade data and their electronic and voice
communications records. Lastly, AI tools can single out alerts that need a more urgent
response.

Regulatory Compliance
Any firm that wants a sound risk management system must comply with all risk
management regulations. To help with compliance, most firms have turned to RegTech
– a subset of fintech that focuses on technologies that can facilitate the delivery of
regulatory requirements more efficiently and effectively compared to the existing
traditional capabilities. AI is an excellent RegTech tool because it allows for continuous
monitoring of firm activities. This way, the firm has access to real-time insights that help
it avoid compliance breaches rather than dealing with the consequences of breaches
after they have occurred.

Potential Benefits of AI and Machine Learning Techniques

in Risk Management
For risk management, AI and machines offer several key benefits:

● Improved data processing: AI and machine learning techniques have made it

possible to process structured and unstructured data in massive amounts.
Datasets can even be combined to form new variables that unravel key
relationships.
● Improved efficiency: Automation of repetitive tasks can help firms to reduce
costs.
● Real-time and predictive insights: AI and machine learning tools can alert firms
about new exposures faster than traditional tools. In addition, machine learning
and AI tools can increase preventative risk advice and help firms develop faster
response times in critical situations.
● Improved decision-making: Machine learning is associated with better
decision-making through greater (predictive) insights and risk visibility.

The Limitations and Challenges of Using AI and Machine

Learning Techniques in Risk Management
Several practical issues plague the use of AI and machine learning techniques in risk
management:

● Availability of suitable data: Although multiple AI and machine learning tools with
the ability to read all types of data have been developed, firms have been slow to
organize their internal data in a way that makes its analysis and processing
simple and straightforward. In some cases, data is held by different departments
(silos) or systems. In fact, sometimes, sharing such data with other internal
departments is restricted. In other cases, firms do not record important data but
merely keep it as “informal knowledge.”
 ● Availability of skilled staff: There’s an overall shortage of skilled employees who
can work with AI and machine learning tools. In the same vein, staff training is a
painstaking, time-consuming exercise that has so far failed to keep up with the
development of new AI solutions. However, there have been efforts to overcome
this problem, notably by creating learning campuses that solely focus on AI and
machine learning techniques. For example, Goldman Sachs has set up a
campus in India aiming to train more than 7,000 individuals on how to use these
tools effectively.
● Accuracy of machine learning solutions: There’s no doubt that AI and machine
learning solutions can help improve risk management. Nevertheless, not all the
underlying tools are effective. Some have been found to suggest unrealistic or
inaccurate solutions that cannot be implemented. As such, continuous monitoring
and constant evaluation have to accompany the use of all machine learning
solutions. Firms have to incorporate human input at various stages instead of
giving algorithms full control (from data gathering to decision-making).
● Transparency and ethics: As good as AI and machine learning tools can be,
there seems to be a consensus that some of the underlying solutions are not as
transparent as firms and regulatory authorities would want. One of the most
contentious solutions is deep learning, where models work out the output in a
black box system. If a firm cannot clearly demonstrate how certain decisions are
made, it would be difficult to convince regulatory authorities that the models
being used are valid and ethically sound.

Question
In the context of the Credit Default Swap (CDS) market, a financial institution is
contemplating the use of artificial intelligence (AI) and machine learning techniques to
manage its credit risk. Which of the following statements is correct regarding the
application of these techniques?

A. AI and machine learning techniques are best suited for evaluating and improving the
liquidity of CDS instruments in the secondary market.
B. AI and machine learning techniques are primarily useful for enhancing customer
interactions in the context of the CDS market.

C. The use of AI and machine learning in the CDS market is most valuable for
improving audit trails and ensuring regulatory compliance.

D. Employing AI and machine learning techniques can improve the prediction accuracy
of credit events and default costs in the CDS market.

Solution

The correct answer is D.

While AI and machine learning techniques can have a range of applications within
financial markets, in the Credit Default Swap (CDS) market specifically, their primary
value stems from their potential to handle the complexity and uncertainty of assessing
credit risk. These technologies, particularly models involving deep learning, can
enhance the prediction accuracy of credit events (i.e., the likelihood of a default) and the
estimation of default costs. This predictive capability is vital for managing risk and
making informed investment decisions in the CDS market.

Options A, B, and C, although plausible applications of AI and machine learning

techniques, don’t directly address the key challenges inherent in credit risk assessment
within the CDS market. Therefore, option D is the most likely correct answer.
Climate-related Risk
Drivers and their
Transmission Channels
After completing this reading, you should be able to:

● Describe climate-related risk drivers and explain how those drivers give rise to
different types of risks for banks.
● Compare physical and transition risk drivers related to climate change.
● Assess the potential impact of different microeconomic and macroeconomic
drivers of climate risk.
● Describe and assess factors that can amplify the impact of climate-related risks
on banks as well as potential mitigants for these risks.

Climate-related Risk Drivers

Climate-related risks refer to climatic changes that could potentially give rise to financial
risks. These risks have increased significantly over the last 100 years due to global
warming, which has, in turn, increased the frequency of extreme weather events. The
result is loss of lives, diminished livelihoods, reduced production in plants and animals,
and damaged infrastructure, among other adverse impacts.

Climate risk drivers can be grouped into one of two categories:

● Physical risks.
● Transition risks.

Physical Risks
Physical risks are tied to weather and climatic changes that impact the economy. They
can be subdivided further into acute risks, which come about due to extreme weather
events, or chronic risks associated with long-term progressive shifts in climate. Acute
physical risks include wildfires, heatwaves, floods, storms, hurricanes, typhoons, and
cyclones. Chronic physical risks include rising sea levels, ocean acidification, and rising
temperatures. Prolonged periods of high temperatures can also lead to desertification.

Physical climate risks may occur with a significant time lag. What’s more, severity differs
from one event to another. Human activity and day-to-day decisions, to an extent, affect
exposure to physical risks. Nevertheless, it’s impossible to control the location, timing,
and magnitude of specific physical events.

Transition Risks
Transition risks refer to societal disruptions arising from adjustments towards a
low-carbon economy. Migration to a low-carbon economy comes with a host of changes
that impact not just working conditions but also the products manufactured.

Sources of transition risk include:

● Changes in public sector policies.

● Innovation and modifications in the affordability of existing technologies (e.g., that
make renewable energies cheaper or allow for the removal of atmospheric GHG
emissions).
● Investor preference for greener tools and resources.

Banks have been caught up in these risks, a situation that has given them the incentive
to deploy risk-monitoring tools in an attempt to mitigate or eliminate risk effects.
However, the sheer scale and synchronous nature of transition-related changes mean it
isn’t easy to keep up, and the impact can be greater than previously anticipated. It is
noteworthy that transition risk drivers vary from one economy to another depending on
the existing levels of technology and mechanization.

Let’s now briefly look at examples of transition risk drivers.

Climate Policies
In recent years, countries worldwide have put a lot of effort toward finding solutions to
risks resulting from climate change. Through the Paris Agreement, an international
treaty that enjoys the support of 191 countries (and the European Union), nations have
pledged to take a host of measures and enact policies that reduce GHG emissions and
adopt low-carbon economies. Some countries have gone as far as barring the
importation of certain products while setting deadlines for the manufacture of certain
local goods. For example, the UK has pledged to reduce greenhouse gas emissions by
at least 100% of 1990 levels (net zero) by 2050.

Technology
There’s been a sustained push around the world to replace old technology with new
technology and tools that emit little greenhouse gases. For example, counties are
encouraging automobile producers to ditch the production of gas-dependent cars in
favor of electric models, which emit less carbon into the atmosphere. The problem is
that some of the proposed technological changes might force companies to ditch proven
long-term business models and adopt the use of resources that may become more
expensive over time. On the upside, firms that are quick to adopt the changes stand to
benefit from public goodwill and favorable government policies. The government can
also impose taxes on the use of certain resources.

Investor Sentiment
Investors are increasingly factoring climate risks into their investment decisions, a trend
that may reflect pressure from non-governmental organizations and environmental
groups. Indeed, some of the world’s largest asset managers are already incorporating
climate change into investment decision-making and investment approaches. For
corporations directly impacted by climate change, both their bond and equity offerings
will be subject to valuation and re-valuation as investors change their assessment of the
underlying climatic risks.

Consumer Sentiment
Adapting to a low-carbon economy requires a change in human behavior.
Climate-friendly consumption would, for example, lead to more climate-friendly
transportation, manufacturing, and energy use. Evidence indicates that there is a shift in
consumer behavior already underway. Clients of retail banks may ask that their savings
or investments be directed to institutions with more eco-friendly policies or projects that
contribute to the environment. A growing awareness of, and explicit demand for,
climate-friendly financial products and investment could spur corporations and banks to
change their business strategies, notwithstanding potential regulatory or supervisory
approaches. In the same vein, investors’ and consumers’ expectations of hazards (e.g.,
flooding), climate policies, or technological changes may lead to changes in their
preferences and consequently impact the price of assets.

Assessing the Potential Impact of Different

Microeconomic and Macroeconomic Drivers of Climate
Risk
There’s a clear link between climate risk drivers and financial risk for banks. The causal
chains linking climate risk drivers to the financial risks the banking sector faces are
known as transmission channels. Put differently, transmission channels present the
avenues through which climate change can expose banks to financial risk. These
transmission channels can either be macroeconomic or microeconomic.

Microeconomic Channels
Microeconomic transmission channels refer to the causal chains by which climate risk
drivers affect various individual counterparties doing business with banks, potentially
exposing banks and the entire financial system to climate-related financial risk. They
include the direct effects of climate change on banks, particularly events that disrupt
operations and the ability of banks to raise funds for day-to-day business. In addition,
they include the indirect effects on name-specific assets such as bonds, single-name
CDS, and equities.

Macroeconomic transmission channels are the avenues through which climate risk
drivers affect macroeconomic factors such as inflation, labor productivity, GDP, and
economic growth. These factors may, in turn, have an effect on the economy in which
banks operate.

The following are various microeconomic transmission channels and the financial risk
they create:
Credit Risk
Households
● Severe weather can damage bank-funded property. This, in turn, increases the
probability of default non-performing asset ratios and lowers bank equity ratios.
● Banks using residential real estate as collateral for mortgages may see their
credit risk increase if such property is damaged by adverse weather or rising sea
levels.

Corporates
● There’s evidence that severe weather events (physical risks) reduce corporate
profitability and potentially increase credit risk to lenders.
● Agricultural entities funded by banks can be hit by high temperatures and
precipitation, leading to low yields and problems repaying debt.

Sovereigns and Subnational Institutions

● Physical risk events may lead to lower tax revenues for sovereigns and
supranational institutions resulting from impaired corporates, reduced household
income, and an overall reduction in output. This, in turn, increases the risk of
default and the loss given default for banks with sovereign and municipal
exposures.

Government Policy
● Transitioning toward a low-carbon economy may lower the productivity and
profitability of corporates, affecting their creditworthiness.

Technological Change
● Carbon-intensive technologies may be subjected to heavy taxation to discourage
their use. Thus, any firm that continues to rely on such technologies may find
itself unable to compete against those that quickly adopt newer, more efficient
technology. Credit-related losses may be higher for banks exposed to companies
that cannot adapt to carbon-neutral economies.

Sentiment
As more and more consumers embrace less carbon-intensive products, producers that
stick with the high GHS emission products may see a decline in sales. As such, banks
exposed to such producers may see an increase in credit risk.

Market Risk
● Physical and transition risks can alter or reveal new information about future
economic conditions that will affect the price and value of financial assets. This
may result in downward price shocks and an increase in market volatility.
● Climate risk could also lead to a breakdown of long-established correlations
between financial assets. This would render hedging methods ineffective and
reduce the ability of banks to manage their market risks.

Liquidity Risk
● Banks’ liquidity risk may be affected directly as a result of climate risk drivers,
either through their ability to raise funds or liquidate assets or indirectly as a
result of customer demands liquidity.
● Households and corporations affected by physical risks may withdraw deposits or
borrow funds to cover recovery and other cash-flow needs. Such actions may put
a bank under unprecedented liquidity pressure for banks.

Operational Risk
Physical Risk
● Banks’ operational ability may be reduced if physical hazards destroy
transportation and communication infrastructure.
● Banks and corporations may also see increased legal and regulatory compliance
risks resulting from transition risks.

Macroeconomic Channels
When considering macroeconomic factors, climate risk is expected to have the greatest
impact on credit and market risks.

● Credit risk: First, there is likely to be a climate-related increase in human

mortality, a situation that may result in reduced labor productivity. Second,
empirical evidence suggests climate risk has pushed the cost of debt up by 117
 basis points in developing countries. This means the affected industries may find
it difficult to recover from disasters and honor their financial obligations with
counterparties, including banks. In addition, increased borrowing costs could
bring about higher taxes, lower government spending, and reduced productivity
levels, all of which may indirectly impact the credit risk for banks.
● Market risk: At this point, there’s little research that seeks to establish how the
interaction between macroeconomic factors and climate-related risks can affect
the market risk for banks. However, some evidence suggests that changes in
government policy might affect the value of assets in certain industries that
significantly contribute to the overall economy.

Factors That Can Amplify the Impact of Climate-related

Risks on Banks
The following factors can amplify the impact of climate-related risks:

Risk Drivers Interactions

Interactions exist across both physical and transition risk drivers. One area where we’ve
seen such interactions is the simultaneous introduction of climate-mitigating policies
(e.g., preference for electric vehicles over those that use gas) and technological
breakthroughs.

Financial Amplifiers
Certain financial amplifiers have the potential to increase the impact of climate-related
financial risks on banks. The materialization of climate-related risks on bank balance
sheets might be amplified by behavioral choices within the financial system and
interactions with the real economy. A good example of a financial amplifier is the
absence of insurance or its unaffordability. If bank-funded assets aren’t insured, damage
caused by climate-related events could result in a loss for the bank and result in a
difficult recovery plan for the borrower.

Multiple Channels
Some risk drivers may impact banks through more than one transmission channel, a
situation that amplifies climate-related financial risks. Notably, the interaction between
microeconomic and macroeconomic transmission channels can worsen an already dire
situation. An example of this would be where a physical risk results in the destruction of
houses, thereby affecting the creditworthiness of a bank’s customers while also
impacting the aggregate credit risk for banks.

Mitigants
Mitigants can mitigate and offset banks’ exposure to climate-related financial risks
through proactive and reactive actions. Proactive actions are the pre-emptive steps
banks take to reduce their vulnerability to climate-related risks. Good examples would
be diversification and strategic asset allocation. A bank might increase investment in
sustainable companies, particularly those that have embraced less carbon-intensive
business practices and technology.

Reactive actions include actions taken as a response to climate risks already embedded
in balance sheet exposures. They include insurance and reinsurance, hedging,
securitization, and asset sales that enable a bank to reduce its exposure to high-risk
assets.

Question
Consider a multinational corporation, EarthSpan Inc., which has its manufacturing units
in a coastal region prone to cyclones and its largest consumer base in a highly
developed country committed to implementing stringent green policies. Given the
company’s present context, EarthSpan Inc. faces an assortment of climate change
risks. As a risk manager, you’ve been asked to analyze these risks. Which of the
following combinations accurately describes the physical and transition risks EarthSpan
Inc. may encounter?

A. Physical risk: Reduction in consumer demand due to changing preferences;

Transition risk: Infrastructure damage due to cyclones.

B. Physical risk: Damage to manufacturing units due to cyclones; Transition risk:

Technological obsolescence due to shifting renewable energy technology.
C. Physical risk: Technological obsolescence due to green policies; Transition risk:
Rising sea levels affecting manufacturing units.
D. Physical risk: Decrease in investor interest due to climate concerns; Transition risk:
Flooding of manufacturing units.

Solution

The correct answer is B.

Physical risks associated with climate change relate to changes in weather and climate
patterns that directly affect economies and organizations. EarthSpan Inc., having its
manufacturing units in a cyclone-prone region, faces a physical risk of infrastructure
damage due to cyclones. This risk, categorized as an acute physical risk, is derived
from extreme weather events.

Transition risks are those that arise from the societal shift towards a low-carbon
economy. These risks encompass changes in technology, policy, and societal behavior
towards a more eco-friendly stance. With its largest consumer base in a country
implementing stringent green policies, EarthSpan Inc. could face a transition risk of
technological obsolescence, as the shift towards renewable energy technology may
outdate their current manufacturing processes.

Option A is incorrect because it incorrectly identifies a reduction in consumer demand

due to changing preferences as a physical risk. In reality, this would fall under transition
risk, as it reflects changes in consumer behavior driven by evolving societal sentiments
and preferences. In the same vein, infrastructure damage due to cyclones is actually a
physical risk as it pertains to the impact of extreme weather events on the organization’s
physical assets.

Option C is incorrect because it misclassifies technological obsolescence due to green

policies as a physical risk. Technological obsolescence is a transition risk, as it arises
from changes in technology and regulations related to the transition to a low-carbon
economy. Rising sea levels affecting manufacturing units, on the other hand, is a
physical risk, as it pertains to the direct impact of climate-related changes on the
organization’s physical infrastructure.
Option D is incorrect because it incorrectly identifies a decrease in investor interest due
to climate concerns as a physical risk. This is actually a transition risk, as it relates to
changes in investor sentiment and their assessment of the organization’s exposure to
climate-related risks. Flooding of manufacturing units, however, is as a physical risk, as
it pertains to the direct impact of climate-related events on the organization’s physical
assets.