FRM II - Current Issues 2020

Cheat Sheet

1-The Impact of Blockchain Technology on FinanceA. Blockchain Technology Basics

i. Blockchain

1. Computerised ledger; cryptographic techniques; consensus and secure.

2. The ledger is distrusting, none single point can control over it.

3. An ever-growing chain of ledger entries links the entire history.

ii. A Brief History of Consensus

1. Distributed Consensus

a. Agree on common data in the presence of faults

b. Blockchain uses distributed consensus to agree upon transactions.

iii. Bitcoin

1. Participants can agree upon a continually updated history of all transactions.

2. Users have control over their bitcoin via a digital signature.

3. Digital signatures are public, cannot be forged, and can be verified by anyone.

iv. The Proof-of-Work Design

1. Competing to find a randomly chosen string of numbers and letters.

2. Once a miner finds, they broadcast it, along with the block and claim their reward.

v. The Adjustment on the Difficulty of the Competition

1. An algorithm automatically adjusts the difficulty of mining.

2. As more miners join the network, the difficulty rises.

3. Miners incur costs in the form of specialised computer hardware and electricity.

vi. Bitcoin Full Nodes

1. Computers only validate the Bitcoin blockchain, but do not mine new blocks.

2. To validate that the blockchain includes the transaction correctly.

vii. Blockchain Technology and Distributed Databases

1. Permissionless blockchains

2. Permissioned blockchains

3. ‘Distributed ledger technology’ (DLT)

viii. Smart Contracts

1. Enforce the transfer of digital assets.

2. Do not require a trusted third party to intermediate.

3. The blockchain network enforces the execution of the contract on its own.

ix. Tokens

1. Companies issue a new token managed by a smart contract.

2. The blockchain network’s computers validate the token’s transactions.

x. A Spectrum of Decentralisation

1. On the decentralised end: permissionless systems

2. On the other end: a traditional centralised database

3. Costs from Introducing a Blockchain

a. Decentralised security is quite costly.

b. A trusted third party to secure the ledger, operating costs are likely to be lower

c. Must weigh the ‘cost of trust’ against the high operational costs.

B. Blockchain Technology’s Opportunities and Challengesi. Impediments to Broad Adoption

1. Benefits

a. Participants don't need to trust a particular person to maintain that record.

b. Open the door to peer-to-peer transactions.

c. Reduce overall friction.

2. Challenges

a. performance and scale;

b. privacy and security;

c. Interoperability among DLTs;

d. governance, updates;

e. real world use cases;

f. collective action problems;

g. public actions, legal framework.

ii. A Framework for Understanding Transaction Costs and Trade-offs

1. The Trade-offs Worth Consideration

a. Trade-offs between centralised market structures vs distributed networks.

b. The financial sector is choosing security and scalability over decentralisation

Most financial sector applications still rely on trusted intermediaries.

2. Basic Economic Properties of Blockchains

a. More users, higher benefits of blockchain application.

b. For permissioned blockchains, reduce overall transaction costs.

c. In digital identity or medical records: centralised systems may be inferior.

C. Blockchain Technology and Financei. Where Could Blockchain Technology have an Impact?

1. Payments

a. The cross-border payments is slow and expensive.

b. Remittances and foreign currency payments were potential application

2. Digital Identity/Know Your Customer

a. Verify numerous data points about their customers.

b. Banks use DLT as the cross-institution source of proof.

3. Primary Securities Issuance

a. Blockchain-based systems can issue corporate loans.

All parties have a shared record of the transaction and any updates.

Auto distribution of cash flows.

4. Securities Clearing and Settlement

a. A shared ledger may enable real-time clearing and settlement.

5. Derivatives Clearing and Processing

a. Many clauses can be coded directly into smart contracts.

Auto execution.

6. Post-trade Reporting

a. Distributed ledgers include a full audit trail for each transaction.

Facilitate post-trade regulatory reporting.

7. Trade Finance

a. Veracity of exporter claims and make letters of credit more available.

ii. Finance Starting at the Centralised End of the Spectrum

1. Permissioned blockchains mitigate the challenges that public blockchains face.

2. Critics

a. Closed systems face a security risk because the validators can collude.

b. Members may limit the innovations challenging their business models.

iii. Crypto-finance

1. Token Sales

a. Have led to a new means of raising capital- ICO

b. Purchasers are bearing risk of the eventual success of the new network.

2. ICOs

a. VCs see ICOs as a new way to fund start-ups.

b. A high failure rate for ICOs, due to a considerable amount of fraud.

3. Cryptoexchanges

a. Enable investors in the tokens to trade.

b. Offer market-making, advisory and custodial services.

4. Regulatory Responses

a. Payment tokens- subject to anti-money laundering (AML) laws

b. Utility tokens- intended to provide digital access to an application or service

c. Asset tokens- subject to prospectus requirements and trading protections

2-FinTech and market structure in financial servicesA. Background and Definitions

i. Financial Innovation Influences Market Structure

1. The emergence of providers of bank-like services may impact markets.

2. The entry of BigTech firms into financial services increased competition.

3. The reliance on third-party service providers may increase over time.

4. A shift to open banking: greater competition.

ii. Key Elements of Market Structure

1. Concentration

2. Contestability

3. Composition

iii. Tech Innovation Affects these Elements

1. Lower barriers to entry.

2. Increase competition.

3. Incumbents: be constrained by legacy IT systems.

4. BigTech firms: possess up-to-date technology and funds, have a competitive edge.

5. Unbundle many services offered by incumbents.

B. Financial Innovation and Links to Market Structurei. Supply Factors – Technological Developments

1. Use of APIs allows different software applications to communicate with each other.

2. Mobile devices expand the availability of financial services.

3. Cloud computing offers advantages.

ii. Supply Factors – Regulation

1. Changes in licensing and prudential supervision frameworks.

2. Ensuring contestability and a level playing field is an explicit policy objective.

iii. Demand Factors – Changing Customer Expectations

1. The digitisation of commerce => more convenient experiences across the services.

2. Younger cohorts may be more likely to adopt FinTech.

3. A desire for higher returns => provide FinTech platforms with a larger investor base.

4. The convenience of investing through mobile.

C. The Current Landscapei. Impact to Date of FinTech Firms

1. FinTech firms have found new niches and underserved clients.

2. The Case of Lending Platforms

a. Competitive pressures on incumbent lenders appear limited.

3. Cooperation between Incumbents and FinTech Firms

a. Incumbents outsource to FinTech firms their business.

b. FinTech firms can access to incumbents’ client base.

ii. Impact of BigTech Firms

1. Client data allows them to carry out risk assessments.

2. Acquire market share in the high-revenue area.

3. BigTech Firms Differ from FinTech Firms

a. BigTech firms already have established networks and a large customer base.

b. Well capitalised.

c. Have ready access to the forefront technologies to process big data.

iii. Third-Party Service Providers

1. The level of reliance on cloud computing providers may be low.

iv. How Firms Utilise Cloud Computing

1. Cloud computing has the potential to improve the security and resilience.

a. Avoid vendor capture.

b. More affordable.

c. Promote greater security.

d. Enable small FIs access to sophisticated architecture.

D. Conclusions on Financial Stability and Implicationsi. Summary of Findings

1. Implications for Financial Stability

a. FinTech firms may partner, complement, or compete with existing FIs.

2. The Benefits for FinTech Firms

a. Compete more effectively in some narrow product areas.

b. Reduce the stickiness of existing customer relationships.

c. Greater decentralisation of financial services.

d. Enhancement of financial stability through wider access to financial services.

3. Macro-Financial Risks

a. The effects of competition.

b. Disruption of business models on profitability.

c. Loosening of lending standards.

4. Micro-Financial Risks

a. Cyber incidents.

b. Heightened legal risk.

ii. Implications

1. Vigilance by Supervisors

a. Monitor the impact of competition on profitability and lending standards.

b. Monitor cyber risk.

c. Understand the incentives and barriers to entry by BigTech firms.

d. Manage third-party risks.

E. China’s NPI’s online MMFsi. China’s NPI’s online MMFs market impact and risk analysis

1. Regulatory arbitrage

2. Pushed up financing costs for banks

3. Potential liquidity risks due to potential maturity mismatches

4. Assuming systemic importance in china

ii. Recent measures targeted at Chinese NPIs’ online MMFs

1. Sold by commercial banks or licensed MMF sale agents.

2. Capping instant T+0 redemption.

3. Prohibiting T+0 redemption with their own cash in advance.

4. Prohibiting NPIs from engaging in the sales of money market funds.

3-Fintech credit markets around the worldA. How does Fintech Credit Work?

i. Fintech Credit

1. Credit activity facilitated by electronic platforms -not operated by commercial banks.

2. Use digital technologies to interact with customers online.

3. Lie outside the prudential regulatory perimeter.

ii. P2P

1. The online platform provides a low-cost standardised loan application process.

2. The platform facilitates direct matching and transacting of borrowers and lenders.

3. The credit platform services the loan in return for ongoing fees.

iii. Risk Spread and Default Risk

1. Spread their investments across multiple loans.

2. Platforms maintain a contingency fund for borrower defaults.

iv. Revenue Sources

1. Retain and attract an investor base to generate fee revenue.

2. Some platforms retain loans on their balance sheet.

v. The Digitalisation of Customers and Loan Origination

1. An assessment of borrowers’ credit quality: a credit grade => set a loan interest rate.

2. Many platforms tend to assess non-traditional data.

vi. How Banks Embrace Fintech Credit?

1. Banks‘ use of new digital techniques is not yet as advanced.

2. Fintech is distinct from lower-yielding, but safer, bank deposits.

B. Fintech Credit Market Developmenti. Market Development

1. China was by far the largest market.

2. But a slowdown in many major jurisdictions.

3. Fintech credit represents a very small share of overall credit flows.

4. Measuring the size of fintech credit is challenging, in part because of its novelty,

small size and diversity

ii. The Composition of Fintech Users

1. Individuals or institutional investors.

C. What Drives Fintech Credit?i. Three major factors

1. Fintech credit volume per capita is positively associated with GDP per capita.

2. A less competitive banking sector => more fintech credit activity.

3. More stringent banking regulation deters fintech credit activity.

D. Implications for Credit Availability and Riski. Benefits

1. Greater convenience, lower transaction costs and better credit risk assessments.

2. Greater financial inclusion.

3. Greater diversity in the sources of credit.

ii. Risks

1. Could weaken lending standards.

2. More procyclical than traditional credit.

3. More vulnerable to investor pullback.

4. Erode incumbent banks’ profitability.

5. Higher platform default rates—Investor confidence has also been shaken.

E. Regulatory Frameworksi. Regulatory Forms

1. Licences to operate fintech credit platforms.

2. Minimum capital requirements.

3. Prohibit some high-risk business models.

4. Mandated filing.

ii. Forms to Encourage Innovation

1. Regulatory sandboxes.

2. Innovation hubs.

3. Funding support.

4. Specific tax incentives.

iii. Impacts on the Supervision of Existing Financial Intermediaries

1. Banks interact with platforms => present new reputational and operational risks.

2. Monitor potential related macro-financial vulnerabilities.

4- Implications of fintech developments for banks and

supervisorsA. Fintech Developments and Forward-looking Scenarios

i. What is Fintech?

1. “Technologically enabled financial innovation that could result in new business

models, applications, processes, or products with an associated material effect on

financial markets and institutions and the provision of financial services”.

ii. What Are the Key Fintech Products and Services?

1. Three Product Sectors

a. Credit, deposit and capital-raising services;

b. Payments, clearing and settlement services;

c. Investment management services.

iii. How Big is Fintech?

1. Fintech has reached the initial peak of the “hype cycle”.

2. But volumes are currently still low.

iv. Comparison with Previous Waves of Innovation and Factors Accelerating Change

1. The pace of adoption has increased.

2. Ageneration of digital natives is growing up with a technological proficiency.

3. The effects of innovation and disruption can happen quickly.

v. Forward-looking Scenarios

1. The better bank: The incumbent banks digitise and modernise themselves.

2. The new bank: Incumbents are replaced by challenger banks.

3. The distributed bank: New businesses emerge to provide specialised services.

4. The relegated bank: incumbent banks become commoditised service providers.

5. The disintermediated bank: Banks have become irrelevant with customers.

6. Future evolutions may be a combination of the five scenarios.

B. Implications for Banks and Banking Systems

i. Opportunities

1. Financial inclusion.

2. Better and more tailored banking services.

3. Lower transaction costs and faster banking services.

4. Improved and more efficient banking processes.

5. Potential positive impact on financial stability due to increased competition.

6. Better Regtech.

ii. Key Risks

1. Strategic risk: rapid unbundling of bank services increases risks to profitability.

2. Systemic operational risk: IT interdependencies cause IT risk into systemic crisis.

3. Idiosyncratic operational risk: innovative products increase the complexity of financial

services delivery.

4. Compliance risk with regard to data privacy.

5. Outsourcing risk

6. Cyber-risk

7. Liquidity risk

iii. Implications of Using Innovative Enabling Technologies

1. AI: gain greater insight into customer needs and provision of more tailored services.

2. DLT developments facilitate value transfer exchanges without intermediation.

3. Cloud Computing

a. Incumbents develop new solutions and migrate away from legacy systems.

b. New players can fit scenarios that challenge the current banking system.

C. Implications for Bank Supervisors and Regulatory Frameworksi. Increased Need for Cooperation

1. Ensure banks using innovative technologies are complying with the relevant laws.

ii. Bank Supervisors’ Internal Organisation

1. Ensure the knowledge, skills and tools of their staff remain relevant and effective.

iii. Suptech Opportunities

1. New technologies to improve the supervision methods and processes.

iv. Continued Relevance of Regulatory Frameworks

1. Supervision of Third-party Service Providers

2. Licensing Regimes

a. Consider new regulations related to emerging fintech services.

v. Facilitation of Innovation

1. Innovation hubs

2. Accelerators

3. Regulatory sandboxes

5-The Rise of Digital MoneyA. New Digital Forms of Money

i. A Taxonomy—The “Money Tree”

1. Four Attributes of Payment Means

a. Type: An object such as cash or a claim such as swiping a debit card.

b. Value: Fixed value claims or Variable value claims.

c. Backstop: Backstopped by the government, or by private?

d. Technology: Is settlement centralised or decentralised?

ii. Five Different Means of Payment

B. Adoption of E-money Could be Rapidi. How Stable is E-money?

1. Cryptocurrency

a. Its value can fluctuate significantly.

2. Central Bank Money

a. Cash or CBDC: is perfectly stable.

b. The government’s solvency underpins the value.

3. i-Money

a. I-money backed by Treasury bills will be less risky than backed by stock.

4. E-money

a. Does not benefit from government backstops.

5. E-money is Exposed to 5 Types of Risks

a. Liquidity, default, market, foreign exchange rate risk and operational risk

6. Options to Minimise Exposure to These Risks

a. Invest in safe and liquid assets.

b. Control the creation of e-money.

c. Sufficient capital.

ii. E-money Adoption Could Be Fast for Its Attractiveness as a Means of Payment

1. Why E-money Grows Rapidly?

a. Convenience.

b. Transaction costs.

c. Users trust telecommunications and social media.

d. Network effects.

e. Ubiquity.

f. complementarity.

C. Effects of E-money on the Banking Sectori. Risks of Rapid E-money Adoption

1. E-money providers may be natural monopolies.

2. Risks to monetary policy transmission could emerge from currency substitution.

ii. Scenario 1: e-money and b-money will coexist; the battle will wage on.

iii. Scenario 2: e-money providers could complement commercial banks.

iv. Scenario 3: commercial banks’ deposit-taking and credit functions could be split.

D. The Role of Central Banks and Synthetic CBDCi. Today’s World

1. All banks hold accounts at the central bank.

ii. Tomorrow’s World

1. Some central banks already offer special purpose licenses that allow nonbank

fintech firms to hold reserve balances.

a. Allow e-money providers to overcome market and liquidity risk.

2. Massive Runs from Bank Deposits into E-money in Times of Crises?

a. As clients seek the protection of banks’ deposit insurance: flow from e-money to

b-money.

b. Uninsured deposits might migrate from banks to e-money providers.

iii. Potential Advantages

1. Benefits from Offering E-money Providers Access to Central Bank Reserves

a. Stability of e-money.

b. Central banks could protect consumers from e-money monopolies.

c. Monetary policy transmission could be more effective.

iv. Synthetic Central Bank Digital Currency

1. The central bank: offer settlement services, access to central bank reserves.

2. E-money providers: all other functions.

3. Dual comparative advantages.

6-Big Data

A.Tools to Manipulate Big Data

i. Why Big Data is Useful?

1. More powerful data manipulation tools

2. Variable selection

3. More flexible relationships than simple linear models

B.Tools to Analyse Data

i. Pre-Processing Data

ii. Categories of Data Analysis

1. Prediction

2. Summarisation

3. Estimation

4. Hypothesis testing

iii. Machine Learning

1. CART, random forests, Neural nets, Deep learning, SVM, etc

C.General Considerations for Prediction

i. Overfitting Problem

1. Several Ways to Deal with Overfitting

a. Simpler models

b. Training, testing, and validation

c. K-fold cross-validation

ii. Works well in-sample but fails miserably out-of-sample

D.Classification and Regression Trees

i. CART

1. To predict a 0 or 1 outcome

ii. Example: Classification for Survivors of the Titanic

1. CART vs Logistic Regression

a. Logistic regression: better for smaller data sets

b. Trees: better for larger data sets

c. Trees not to work very well if the underlying relationship is linear

2. Pruning Trees

iii. Example: Home Mortgage

1. Bootstrap: choose with replacement a sample

2. Boosting: repeat estimation

3. Bagging: average across models

4. Random Forests

a. a black box

E.Econometrics and Machine Learning

i. Collaboration between Econometrics and ML

1. Time Series Models

2. Causal Inference

a. Machine learning dealt with pure prediction

b. Econometricians: causal inference

3. Causality and Prediction

ii. Estimating the Causal E ectff of Advertising on Sales

iii. Model Uncertainty

1. Averaging over many small models

7-Machine Learning

A. Introduction

i. Machine Learning and Artificial Intelligence

1. More details of reporting

2. High-frequency, unstructured consumer data

3. ML: model complex, non-linear relationships

B.Background to ML

i. Supervised ML

1. Dependent variables are known

ii. Unsupervised ML

1. A corresponding dependent variable is lacking

iii. Non-parametric Analyses

1. Flexible to fit any model

2. Hardly make an assumption about the relationship

3. Infer non-linear relationships

iv. Machine Learning Methods

1. Regression

a. A supervised ML problem.

b. Continuous dependent variable

2. Classification

a. A discrete problem

3. Clustering

a. An unsupervised ML problem

v. Prediction versus Explanation

1. Statistical methods are good for explanation

2. ML is good for prediction

vi. Tackling Overfitting: Bagging and Ensembles

1. Overfitting

a. Perform poorly when tested out-of-sample

b. Having too many parameters

2. Ways to Deal with Overfitting

a. Boosting and bagging

b. Random forest

c. Ensemble

i. Averaging over many small models A

vii. Theory-free Approach to Analysis?

1. ML: not understanding of the relationship Deep

viii. Learning and Neural Networks

1. Deep Learning

ix. Application within Financial Services

1. Massive Data

a. High-quality, structured supervisory data

b. High frequency, low quality “big data”

C.Four Use Cases

i. Credit Risk and Revenue Modeling

1. Application

2. Di culties inffi Usage

a. Models can be sensitive to overfitting the data.

b. Hard for any human to understand.

ii. Fraud

1. Detection of Credit Card Fraud

a. Clear historical data with relevant fraud labels to train classification

iii. Detection of Money Laundering and Terrorism Financing

1. Current Challenges

a. Unable to detect complex patterns of transactions

b. Significant human capacity is required

c. Impediments to data sharing

2. ML’s Role

a. Money laundering is hard to define

b. Unsupervised learning

iv. Surveillance of Conduct and Market Abuse in Trading

1. Application

a. Monitor the behaviour of traders

2. Challenges to Applying ML

a. No labeled data to train algorithms

b. Black boxes: hard to explain to a compliance o cer ffi

3. Barrier to the Implementation of Automated Surveillance

a. Information from di erent sources could be mutually incompatibleff

8-Artificial intelligence and machine learning in financial

services

A. Background and Definitions

i. AI and ML

ii. Categories of ML Algorithms

iii. Problems of Causality

B. Drivers

i. On the Supply Side

1. Benefited from AIML tools developed for other fields

a. Hardware costs, computing power, algorithms, storage...

2. Infrastructure construction

a. E-trading platforms, social media, credit scoring...

ii. On the Demand Side

1. Cost reduction, risk management gains

2. Regulatory compliance

iii. Impact future adoption

1. Growth in the number of data sources

2. Enhanced data quality

3. Improvement in hardware / software

4. Open source libraries

iv. The Legal Framework

1. Breaches of personal data

2. New data standards

C. Selected Use Cases

i. Customer-focused Uses

1. Credit Scoring Applications

a. Speed up lending decisions

b. Enable greater access to credit

c. A/D to Using AI in Credit Scoring Models

Lack transparency Bias

2. Use for Pricing, Marketing and Managing Insurance Policies

a. Improvements to the underwriting and claims processing

b. Consumer protection concerns

3. Client-facing Chatbots

a. Giving advice and prompting customers to act

b. Gain information about customers

ii. Operations-focused Uses

1. Capital Optimisation Use Case

a. Better capital optimisation

2. Model Risk Management and Stress Testing

a. Back-testing and model validation

b. Stress testing

3. Market Impact Analysis

a. Assess the market impact of a given trade

b. Minimise trading impact

c. Identify the timing of trades

iii. Trading And Portfolio Management

1. AIML in Trading Execution

a. Improve the ability to sell to clients

b. Help compliance with trading regulations

2. Scope for the Use of AIML in Portfolio Management

a. More e ective use of the vast amount of availableff data

b. ML can only generate predictive power

iv. AIML In Regulatory Compliance And Supervision

1. RegTech: Applications by Financial Institutions for Regulatory Compliance

a. Facilitate regulatory compliance

Interpret data inputs such as e-mails

Interpret regulations into a common language

KYC (know your customer)

2. Uses for Macroprudential Surveillance and Data Quality Assurance

a. Automate macroprudential analysis

3. SupTech: Uses and Potential Uses by Central Banks and Prudential Authorities

a. Assist with monetary policy assessments

Help authorities to detect

4. Uses by Market Regulators for Surveillance and Fraud Detection

a. Fraud and AML/CFT detection

D.Micro-financial Analysis

i. Possible E ectsff Of AIML On Financial Markets

1. Improvement

a. Collect and analyse info on a greater scale

b. Lower trading costs

2. Concerns

a. Similar AIML programmes => correlated risks

b. Could be used by insiders to manipulate market

ii. Possible E ectsff Of AIML On Financial Institutions

1. Benefiting System-wide Stability

a. Increasing revenues and reducing costs

b. Earlier and more accurate estimation of risks

c. Collaboration between financial institutions and other industries

2. Drawbacks

a. Miss new types of risks

b. Black boxes in decision-making

c. For intermediaries: a lack of clarity around responsibility

d. Third-party dependencies

iii. Possible E ectsff Of AIML On Consumers And Investors

1. A Number of Benefits

a. Consumers: lower fees and borrowing costs

b. Wider access to financial services

c. Facilitate more customised financial services

2. Concerns

a. Data privacy and information security

b. Avoiding discrimination

iv. Current Regulatory Considerations Regarding The Use Of AIML

1. Be consistent with the firm’s internal policies and procedures

E.Macro-financial Analysis

i. Potential to Enhance the E ciencyffi of the Economy

ii. Market Concentration And Systemic Importance Of Institutions

1. Degree of concentration

a. A small number of advanced third-party providers

b. Technologies a ordableff only to large companies

2. Reduce the systemic importance of large universal banks

3. Universal banks’ vulnerability to systemic shocks

iii. Potential Market Vulnerabilities

1. Greater diversity in market movements

2. Less predictable trading algorithms

3. Increase liquidity

4. More e ective hedgingff strategies

5. Reduce reliance on bank loans

6. Minimise capital leads to more risk

iv. Networks And Interconnectedness

1. Greater interconnectedness in the financial system

2. Help to share risks

3. Spread the impact of extreme shocks

v. Other Implications Of AIML Applications

1. Reduce the degree of moral hazard and adverse selection

2. Higher premiums for riskier consumers

3. Entail biases

4. ‘Game’ regulatory rules

9-Climate Change and Financial RiskA. Introduction

i. The Early Days of Climate Change in Finance

1. Climate change was not a concern for the financial sector.

2. The rise of carbon markets increases the knowledge about climate change.

ii. An Important Issue for ‘Responsible Investors’

1. Socially responsible investment (SRI) put climate change on the agenda.

iii. The Acceleration

1. Since the 2015 COP21 in Paris.

iv. Climate Policy Meets Finance

1. The Paris Agreement

a. Fully decarbonise the global economy by 2050.

b. Need capital => the role of the financial system is acknowledged.

B. Climate Change as a New Source of Financial Riski. New Types of Financial Risks

1. Transition Risks

a. The financial risks coming in the context of climate change mitigation.

b. Such as climate policies, technology evolutions, consumer preferences.

2. Physical Risks

a. Initial effect

Such as temperature rise, variations in rainfalls.

Affect the operations of organisations.

But the impacts are delayed in the future.

b. Indirect Impacts

Modifications of the climate regimes: heatwaves, droughts, floods...

Related impacts: fires frequency, sea level...

Acute risks: cyclone, flood.

Chronic risk: see level rise.

ii. Are Those New Types of Risk Priced by Markets?

1. Not currently captured, mis-priced by financial markets.

a. Unprecedented Phenomena: No record on how to react.

b. Radical Uncertainty: Unmeasurable, unquantifiable, and uninsurable.

c. Non-normal Probability Distributions

d. Bounded Rationality: May lead to a collective misread of the reality.

e. Discrepancy in Time Horizons: Impacts will only happen in several decades.

f. Climate Change Inefficiently Priced by Markets

C. The Approaches to Manage Climate-related Financial Risksi. Materialisation Channels of Climate-related Financial Risks

1. CC can modify a firm's financial performance and risk profile.

2. Propagation chain becomes more complex if company is multinational diversified.

ii. Climate Scenario Analysis

1. Why Scenario Analysis?

a. A what-if analysis.

b. Avoid attributing scenarios a probability of occurrence.

c. Lie in the comparison of different possible futures.

2. The Difficulty for Scenario Analysis

a. How to translate the factors of climate risks into financial variables?

D. Climate Change Risks and Financial Regulationi. Reporting and Disclosure of Climate-related Risks

1. The Article 173 of the French Energy Transition Act

a. Require reporting on CC.

b. The implementation was unsatisfactory.

2. Mark Carney’s speech

a. Pointed out self-regulation via risk disclosure.

ii. Beyond Reporting, an Enhanced Prudential Framework

1. The European Commission Sustainable Finance Action Plan

a. Mobilise central regulation for CC.

b. Potential methods: through decreasing or increasing capital requirements.

2. Three main objectives:

a. reallocate capital cash flows toward sustainable investments,

b. protect financial system stability

c. promote transparency and long-term visions in financial and economic.

10- Beyond LIBORA. Desirable Features of Reference Rates and Main Trade-offs

i. An Ideal Reference Rate

1. Not susceptible to manipulation.

2. Derived from actual transactions in liquid markets.

3. Serve as a benchmark for both term lending and funding.

ii. The Practical

1. LIBOR Fails to Meet the Criterion

a. Constructed from a survey of banks reporting.

b. Sparse activity in interbank deposit markets.

c. The dispersion of individual bank credit risk.

2. The New Reference Rates Should Incorporate those Attributes

a. Shorter tenor

b. Interbank markets add non-bank wholesale markets.

c. Drawing on secured transactions.

iii. Trade-offs

1. O/N RFRs + Risk Premium

a. O/N RFRs will form the backbone.

b. Another rate embeds a credit risk component.

B. A Taxonomy and Properties of the New Overnight RFRsi. Alternative RFRs in Five Currency Areas

ii. Basic Characteristics of Overnight Reference Rates

1. Volumes underlying new benchmark dwarf those of overnight bank funding rate.

C. Developing RFR-Linked Financial Markets and Term Ratesi. Users of derivatives markets have been accustomed to using compounded O/N rates.

ii. Cash markets have been accustomed to using rates set for the entire term.

iii. State of Market Development

1. IBOR-linked business is still dominant.

iv. Towards Term Benchmark Rates

1. Backward-looking Term Rates

a. Constructed from past realisations of O/N rates.

Less prone to volatility.

Do not reflect expectations about future interest rates.

Lag the actual movements in the O/N rate.

2. Forward-looking Term Rates

a. Known at the beginning of the period.

b. Embed market participants’ expectations.

3. Forward-looking Term Rates Based on Term Funding Instruments

a. Capture fluctuations in intermediaries’ actual term funding costs.

4. Forward-looking Term Rates Based on Derivatives

a. Reflect the market-implied expected path of future O/N rates.

b. Do not capture fluctuations in intermediaries’ term funding risk.

D. Implications for Banks’ Asset-Liability Management

i. A Benchmark for Term Funding and Lending

1. Banks will still lack a benchmark that reflects their marginal funding costs.

2. Banks could be exposed to basis risk.

ii. In Search of Credit-sensitive Term Benchmarks

1. Unsecured Wholesale Funding Rates

a. Unsecured term funding: bank and non-bank.

2. A “Two-Benchmark” Approach

a. Complement the RFRs with reformed local IBOR-type rates.

b. Addresses the scarcity of underlying term transactions.

E. Transition Issuesi. The most pressing thing: the migration of legacy LIBOR-linked exposures to the new

benchmarks after 2021.