Project and IT development

We provide technical and functional expertise, implementation and testing of your traditional or innovative projects.
 

Project management

We always keep in mind that a pragmatic and flexible approach is the best way to meet your needs 

You can base your project on the methodologies we used and our organizational models 

Our consultants provide project services based on 

• The Waterfall method
• The Agile method
• And other methodologies such as Prince2 or Lean
   

Our project management consultants support:

• The organization of the project 
• The preparation of the project
• Writing the business case
• The identification and monitoring of business or technical needs.
• The development of solutions for the realization of complex systems. They establish business and technical constraints, milestones and alerts
• The coordination
• The communication of the planning and of the major events to all stakeholders.
• Delivery management, test organization and production start-up.
• Team management, coaching of internal and external teams. They encourage the self-organization and empowerment of your teams.
• Leading project and steering committees, management committees as well as producing reports for the various stakeholders or sponsors.

Business analysts and architects

The diversity of our technical and functional skills helps us make a success out of your project!
 

Our technical and functional experts act as:

• Business Analyst
• Business process analyst.  This is the analyst who creates process models and workflows.
• IT Business Analyst. It is the professional who is generally involved in needs analysis and problem solving using information technology solutions. He plays a hybrid role, half business, half IT.
• Functional architect. While the job of a Business Analyst is to go into detail, the functional architect tends to remain at a higher and more conceptual level of application. 
• Technical Architect: he works with the functional architect to design the functional behaviour of a system.
   

IT Development

The requirement in your developments! The requirement in the recruitment of our developers!
 

Our developers undergo a series of technical tests before they introduce themselves to you

Our developers have a certain autonomy. We trust them. 

They are allowed to capitalize on their know-how. But they can also rely on a team to advance the projects and improve their skills.
 

Tests and delivery quality 

We consider this to be the most important phase because the successful completion of your project depends on it!
This is a phase that is often overlooked by our competitors. On the other hand, we require our employees to be trained in software testing, and be able to work in a testing environment in order to meet the highest quality standards and thus satisfy you.

Our testers are capable of: 

• Understanding the technical and functional aspects
• Understanding the interdependencies of the recipe of a software project
• Developing, organizing test cases and test scenarios
• Documenting and managing tests
• Carrying out the tests until the acceptance report

Our organization is based on a better understanding:

• The project context: 
  • Understanding of the roles on the project (project owner, project manager, acceptance team, users)
  • The project life cycle: classic or agile modes
  • Deliverables to be carried out (specifications, acceptance plan, tests, etc.)
  • Review of the studied specifications and their business requirements
• The design, formalization and architecture of the tests: 
  • Traceability of requirements and tests
  • Coverage of test objectives
  • Contribution of functional specifications
  • Manual test
  • Automated testing: principles, advantages and disadvantages
  • Functional test robots 
  • Test review 
  • Implementation of tests 
     
• The execution of tests: 
  • Preparation of the campaign
  • Delivery of the version under test and its documentation
  • Testings
  • Recording of results and anomalies
  • Corrective versions / Consolidation of results
  • Stop and acceptance criteria (Go/No go)
     
• The realization of the acceptance phase: 
  • The acceptance phase process and its interactions
  • The steps of the acceptance phase and versioning
  • Non-regression tests on the final evaluation
  • The acceptance report

Risk and regulation: Increasing expectations

We help you achieve your objectives in terms of regulatory risk management and risk governance.
 

BASEL I-I-III and IV
 

The implementation of the Banking Package, consisting of significant amendments to the Capital Requirements Regulation (“CRR”), the Capital Requirements Directive (“CRD”), the Bank Recovery and Resolution Directive (BRRD) and the Single Resolution Mechanism Regulation (SRMR), will occur in the next years

BASEL II-III  to BASEL IV / (CRR and CRD IV) to (CRR2 and CRD V) to (CRR3 and CRD VI)

Rules have been created to implement the Basel Committee's principles for strengthening the resilience of the banking sector by improving the quality and quantity of capital and introducing new liquidity and leverage ratios.

They aimed to adopt certain rules to the complexity, size and profile of the bank in order to avoid a disproportionate application of prudential requirements.

The Basel framework on capital regulation has come to be established as the core assessment of soundness and stability of the banking system The framework continues to evolve with various revisions in response to the changing circumstance, such as the 2008 financial crisis

The previous reforms aimed to complement the current reforms with new measures of which implementation would strengthen the banks' ability to withstand possible shocks. It led to the amendment of several European texts, including the CRR Regulation and the CRD Directive.

The CRD Directive and the CRR Regulation were adopted by the Council of the European Union.

Basel was adopted by Directive 89/647 of 18 December 1989,

Basel II was adopted by Directive 2006/48 of 14 June 2006.

Basel III was adopted by Directive 2013/36 (known as the CRD IV Directive) and Regulation 575/2013 of 26 June 2013 (more commonly known as the CRR Regulation), the latter having been amended by Regulation 2019/876 of 20 May 2019 (the CRR2 Regulation), in order to take account of standards not initially introduced into European legislation.

Basel III also marked a significant tightening of prudential requirements. The amount of the overall solvency ratio has not been significantly changed. On the other hand, the composition of capital was reviewed, in particular in order to eliminate those that had not played the role expected of them during the 2008 crisis.

In addition, the LCR (Liquidity Coverage Ratio), which covers liquidity risk, has been introduced into European prudential regulations.

Basel IV, also defined as Basel III finalized, is for 2022.

The Basel-4 encompasses reforms proposed through consultative documents by the Basel Committee on Banking Supervision (BCBS) at the bottom of the diagram. These reforms address the risks identified in pillar 1 (at the top of the diagram).

Basel-4 is a refinement or reform of Basel-3 but is ultimately very different.

Basel-4

• Deals with only two of the three pillars of Basel: Pillars 1 and 3 (not the second). The bulk of the Basel-4 reform is actually about Pillar 1.
• Another difference is that while Basel-3 focuses on "capital", Basel-4 focuses on risk-weighted assets (RWA).

The BASEL4 reforms

The diagram above describes what the Basel-4 reforms cover.

• No reforms for Pillar 2.
• Disclosure requirements fall under Pillar 3, and the rest under Pillar 1. This includes reforms on Capital Floors; Credit Risk; Counterparty Credit Risk for Derivatives; Securitization; Market Risk; Operational Risk Fundamental Review of the Trading Book; review of the CVA (Credit Value Adjustment) Risk Framework; revisions to Operational Risk; Step-in Risk; and Interest Rate Risk in the Banking Book).

The applications:

EQUITY CAPITAL

• Solvency ratios

• Weighting mechanism.

• Content of accounting equity capital and transition to prudential equity capital

• Components of Equity capital: Common Equity Tier 1 Capital (CET 1), Additional Tier 1 capital (AT1) and Tier 2 Capital (T2 capital).

• Regulatory adjustments.

• Minimum ratio; capital buffers

CREDIT RISK

• Standard Approach and IRB Approach

• Expected losses and depreciations / provisions

  • Provisions and related impairments: The impact of the recognition of "expected credit losses" upon origination or acquisition of assets (IFRS 9 rules applicable from 2018).

  • The treatment of the insufficiency or excess of impairment losses/provisions (compared to expected losses) through the adjustment of Equity capital.

• Securitization rules

• Counterparty risk and CVA

LEVERAGE RATIO

• The application of a binding leverage ratio with a minimum of 3% (Minimum level of equity determined on the basis of total on-balance sheet and off-balance sheet assets calculated without taking into account risk weights)

LIQUIDITY RATIOS

• The Application of the long-term liquidity ratio, the NSFR and the short-term liquidity ratio LCR

MARKET RISK

• Standard Approach and IRB Approach
• Perimeter
• Value at risk
• Fundamental Review of the Trading Book (FRTB)

TLAC ET MREL

The requirement for systemic banks must hold a minimum level of capital and other instruments to absorb losses in the event of resolution  

• TLAC: Total loss absorption capacity
• MREL: Minimum requirement for own funds and eligible commitments (A revision of the rules for setting additional own funds requirements)

INTEREST RATE RISK FRAMWORK (IRRBB)

• Notions of EVE (Economic Value of Equity), Net interest margin
• Stress tests

HIGH RISKS /LARGE EXPOSURE

• GCC (Related Customer Group) notions
• Determination and limit of large risks.
• High Risk Management Framework and link to Risk Appetite Framework (RAF).
• high risk identification and monitoring framework with CRR2.

PILAR II and III

• SREP (supervisory review and evaluation process).
• RAF (Risk Appetite Framework)
• ICAAP (Internal Capital Adequacy Assessment Process).
• ILAAP (Internal Liquidity Adequacy Assessment Process).
• Stress tests.

CRR2 AND CRD V

The revised Capital Requirements Directive and Regulation, commonly referred toas CRD 5 and CRR 2, refine and continue to implement Basel III in the EU by making important amendments in a number of areas

The final CRR2 and CRD5 framework complements and builds on the existing CRD 4 and CRR Regimes with a number of important changes.

• Revised market risk framework (FRTB)

• Standardised Approach for Counterparty Credit Risk (SA-CCR)

• Net Stable Funding Ratio (NSFR)

• Revised leverage ratio requirement

• Revised large exposures framework

• Revised Pillar 2 framework

• Revised regulatory reporting and Pillar 3 disclosure

• Intermediate EU parent undertaking rule

FRTB

Compared to the Basel framework, CRR 2 introduces more precise rules with respect to the proportionate treatment of market risk exposures. This means that the impact of FRTB on larger firms will be more significant while firms with small and medium sized trading books will be subject to a more favorable treatment.

SA-CCR

In line with Basel framework, CRR II adopts a new SA-CCR, which is a more risk sensitive measure of counterparty risk reflecting netting, hedging and collateral benefits, as well as being better calibrated to observed volatilities. The final framework also adopts a simplified SA-CCR and retains the Original Exposure Method (OEM) for smaller firms.

NSFR

In line with the proposals, CRR 2 NSFR framework deviates from the Basel NSFR regime by introducing a number of EU-specific adjustments to make the rules more proportionate for small and non-complex firms.

LEVERAGE RATIO

CRR 2 broadly reflects the Basel leverage ratio. It sets the Tier 1 capital-based leverage ratio requirement at 3% for all EU banks as per the EBA’s recommendation. The final framework confirms that firms are allowed to use any Common Equity Tier 1 (CET1) capital that they use to meet their leverage ratio requirements to also meet their Pillar 1 and Pillar 2 capital requirements.

LARGE EXPOSURES FRAMEWORK

The revised Basel framework defines a large exposure as exposures to a single counterparty that are equal to or above 10% of firms’ Tier 1 capital. The current and future CRR regime prohibits exposures exceeding 25% of firms’ eligible capital.

The final CRR 2 large exposures framework is broadly in line with the proposals and the revised Basel framework, tightening the definition of capital used to calculate the large exposure limit by excluding Tier 2 capital.

PILLAR 2 FRAMEWORK

In line with the initial proposals, CRD 5 revises the Pillar 2 capital regime in the EU legislation. The final framework clarifies the rules around the Supervisory Review and Evaluation Process (SREP) and introduces some limitations on the National Competent Authorities’ discretion when imposing additional reporting and disclosure obligations under Pillar 2.

REGULATORY REPORTING AND PILLAR 3 DISCLOSURES

Unlike the Basel framework but in line with the CRR 2 proposals, the final regime introduces less onerous reporting requirements for small and less complex firms to reduce their compliance burden.

INTERMEDIATE EU PARENT UNDERTAKING (IPU) RULE

Non-EU banking groups (“non-EU groups”) with large EU operations will be required to establish an EU intermediate parent undertaking (“IPU”) according to the final changes to the Capital Requirements Directive (“CRD 5”) package

Our resources are trained to help you calculate your capital needs!

Risk and regulation : Our Services

In recent years, regulatory developments accelerated, business conditions remained uncertain and pressures on management teams and risk and compliance managers increased. As a result, financial companies have been confronted with the impact of new regulations, while struggling to maintain growth and improve their risk and capital management practices and processes
 

Our services are therefore linked to:

• The structure and management of capital.
• The implementation of Basel III and IV regulatory request
• The credit risk, market risk and liquidity risk
• The risk measurement and analysis
• The implementation of new risk related technologies.
• The risk models and their validation
   

The strengths of our teams

• A combination of experience in business processes, risk management, regulation and the financial services industry 
• End-to-end risk management services. Our clients rely on our advice and experience, our experienced team of governance, regulatory and risk specialists, and our ability to provide services anywhere, anytime.
• In-depth technical experience in all disciplines, allowing our experts to objectively understand each situation, identify its specific opportunities and challenges and design an approach that can effectively help clients address these opportunities and challenges.
• An intelligent approach to risk to identify, assess, select and implement resolution strategies
• A practical understanding of internal controls and regulatory compliance processes through a group of experienced professionals
• A high quality analytical capacity through project management tools
• A current knowledge of market trends and emerging technologies.

We help your company meet regulatory requirements

ICAAP, ILAAP, SREP

ICAAP :

The role of ICAAP is to assure that a company has a robust wide risk management framework, that its business model is viable and sustainable, and also to identify and assess all material risks, control the amount and quality of internal capital in relation to its risk profile and to validate the capital adequacy.

Our teams commit to developing an analysis strategy on ICAAP

• They collect and analyze data about current methodologies;
• They review old models and the development of new models;
• They assist in business plan and financial forecasts;
• They perform an assessment of the risk based on the capital;
• They review the calculations and simulations to quantify and aggregate risks, including risk-type level stresses and stressed capital requirements;
• They write a documentation of the results;
• They compute the capital adequacy;
• They determine the capital requirements;
• They assess the impact of the stress test;
• They review  the ICAAP through the governance process;
• They write a final ICAAP report.

ILAAP :

The role of ILAAP is to have a clear understanding of its liquidity characteristics, assess its liquidity risks from an asset and funding illiquidity perspective. ILAAP assesses the appropriateness of governance and controls to identify manage liquidity risks and validate liquidity adequacy.

Our teams work to develop an analysis strategy on ILAAP

• They conduct a detailed review of the bank’s liquidity risk management framework to identify any inconsistencies/gaps with Basel Principles, regulatory guidelines;
• They review key liquidity risk drivers and liquidity stress testing scenarios, identifying any required enhancements;
• They assist in the assessment of adequacy of risk appetite and contingency funding plan;
• They assist in the assessment of supporting rationale for key liquidity metric thresholds and limits;
• They write a document covering recommendations for enhancement of the quantitative liquidity framework and stress scenarios to determine liquidity buffer;
• They write a final ILAAP Report.

SREP and pillar II :

Under the EBA guidelines, the scope of the SREP has considerably widened to include many other elements such as business model analysis, Individual Liquidity Adequacy Assessment Processes (ILAAPs), internal governance and controls.

We assist you in developing and integrating your ICAAP and ILAAP process along with the new practices.

• Integration with the business strategy

SREP is a common framework and methodology used by regulators to assess institutions’ risks and viability. Institutions are expected to have their ICAAP and ILAAP processes intertwined with the business strategy of the Bank. Banks’ risk appetite and stress scenarios reflect the business model; parameters and results emanating from the ICAAP and ICAAP processes should be important resources for business decision making. ICAAP and ILAAP could therefore be used as management tools instead of mere regulatory documents.

The goals are to :

• Acquire a clear understanding of its strategic vision and business model.

• Have appropriate documentation that supports the business model’s articulation.

• Ensure that the board and senior management have a consistent view of the Bank’s business model.

• Ensure that key dependencies and vulnerabilities related to the business model are identified.

• ILAAP is a new assessment process

Within the SREP related to the implementation of CRR/ CRD IV (‘Basel III’) the ILAAP is introduced as a new assessment process, complementary but distinct to the ICAAP, intended to strengthen the robust management of liquidity risk.

• Governance

Institutions should produce, at least once per year, a clear formal statement on their capital adequacy supported by an analysis of ICAAP outcomes, approved and signed off by the management body.

• ICAAP perspective

Institutions are expected to implement a proportionate ICAAP approach with their own survival in mind.

• Risks considered

Institutions are responsible for implementing a regular process for identifying all material risks, as shown opposite.

• Assumptions and key parameters

These should be set in line with risk appetite, market expectations, business model and risk profile.

• Severity levels of stress tests and scenarios have to be tailored to the institution’s vulnerabilities resulting from its business model and operating environment. At least once a year, institutions shall perform an in-depth review of their vulnerabilities. We help develop a stress test plan and methodology adapted to the size and complexity of the institution, by designing models and scenarios, selecting macroeconomic factors, measuring sensitivities and developing prospective positions. This involves understanding the requirements of stress testing, defining a methodology, mapping flows and identifying controls and limitations.

RWA optimization :

Many banks are looking for advanced approaches to optimize their capital requirements and improve the risk sensitivity of their RWA calculations. Lfzpartners helps large institutions to identify, prioritize and implement RWA optimization and improve their credit and market risk calculation processes.

CREDIT RISK

Further granularity is provided for unrated exposures (an 85 percent risk weight if the exposure is to a corporate SME), and for Specialized Lending (where the risk weight can range from 80 percent for high quality project finance to 150 percent for exposures rated below BB-). As with residential real estate, the changes appear to hit hardest certain “niche” lenders where the characteristics of the exposure can be associated with a higher risk volatility.

Under the new CR-SA some of the exposures in the risk weight table must be recalibrated, and there is a more granular approach applied when the exposure is unrated (or when ratings are not permitted for regulatory purposes)

The revised CR-SA will apply from 1 January 2022, along with the changes to other risk types and the initial phase in of the output floor.

IRB METHOD

The new CR-IRB

• Restrictions on the IRB approach
  • is no longer allowed for exposures to banks and other financial institutions, or for corporates belonging to a group with total consolidated annual revenues greater than €500 million (note that Foundation- AIRB is still allowed for these exposures).
  • No IRB approach is allowed for equity exposures

• Risk Weighted Asset (RWA) calculation – removal of the 1.06 scaling factor used in the calculation of Risk Weighted Assets for credit risk exposures.

• Risk parameter floors: Introduction  of PD, LGD, EAD and CCF floors for corporate and retail exposures. For corporate exposures the minimum PD (floor) has increased from 0.03 percent to 0.05 percent, and LGD floors set for different collateral types. Similarly, new PD and LGD floors are in place for retail exposures

These changes, in particular the restriction on the IRB approach and the introduction of parameter floors, will have a direct impact on Pillar 1 capital requirements.

TLAC MREL

Due to size, complexity, and interconnectedness banks cannot undergo regular insolvency proceedings in the context of various financial crises.

The consequences of massive bank failures are deemed unpredictable.

The main objective of the following regulatory initiatives is to ensure that possible future bank resolutions are feasible and realistic, without the government and ultimately the taxpayer having to cover losses, regardless of the institution’s size.

The "Principles on Loss Absorption and Recapitalization Capacity" (also referred to as "TLAC") introduced the concept of an additional liability requirement for the largest institutions.

The key element of this concept is the so-called "bail-in" tool, which aims to require the institutions' creditors to share in the institution's losses and thus banish the danger of another public bail-out.

MREL, TLAC and Bank Resolution : In Europe, the Bank Recovery and Resolution Directive (BRRD) provides the legal basis for bank resolution procedures, resulting in an entirely new regulatory framework.

The BRRD introduced an approach similar to TLAC, but which can be applied to virtually all institutions in the European Union, not just systemically important institutions : the Minimum Capital Requirement and Eligible Liabilities (also known as MREL).

Both TLAC and MREL focus on the same key aspect : increasing the loss-absorbing capacity of the banking sector through a binding minimum ratio for loss-absorbing liabilities.

The implementation of the resolution regime affects institutions to a large extent : Therefore, banks must be able to provide a clear and transparent picture of their business functions and report it to the resolution authorities.

IRRBB

Interest rate risk - Definition :

Interest rate risk can be summarized as IRRBB (Interest Rate Risk in the Banking Book) and the estimation of EV (Economic value) and NII (MNI) which are the main indicators of interest rate risk

Definition

The concept of IRRBB (Interest Rate risk Banking Book) is introduced here

 "When interest rates change, the present value and timing of future cash flows change, thereby altering the underlying value of assets, liabilities and off-balance sheet items and hence the economic value of the institution. Changes in interest rates also affect earnings by altering interest sensitive income and expenses, which in turn affects net interest income. If not managed properly, excessive IRRBB can pose a significant threat to a bank's capital base or future earnings.’’

Several measures of interest rate risk exist and include

The risk reduced to the bank's own funds following an instantaneous rate shock.

The Outlier Test ratio can be used directly. This is a static indicator because balance sheet operations are taken into account without renewal and a NPV (Net Present Value) on which a variation of +/- 200bp is applied, other deformations of the rate curve can also be considered to which we will introduce the notion of EVE

Risks due to variations in interest rates Impacting future profits (MNI): (Sensitivity calculated from several rate shocks, balance sheet operations are renewed here, the impact is measured in relation to a sensitivity between the current scenario and the most unfavorable scenario)

Historical 

Interest rate risk in the banking book or "IRRBB" is part of the second pillar of Basel Capital and has become a recent concern of the EBA.

The Basel Committee presented two solutions in 2015, Pillar 1 (Minimum Capital Requirements) and Pillar 2. Given the practical difficulty of banks applying Pillar 1, the Committee decided that Pillar 2 is the most suitable for the treatment of IRRBB.

The EBA launched a consultation paper on 31 October 2017 and seems to agree with the Basel Committee (interest rate risk included in Pillar 2 and confirmation of interest rate risk indicators such as the outlier test or the EVE) while insisting on the capitalization of interest rate risk in Pillar 2)

Interest rate risk - Focus on the outlier test :

OT (Outlier test)

The OT is a measure of static interest rate risk (e.g. the sensitivity of van can be calculated in relation to several interest rate shocks including +/- 200 bps);

This indicator is often based on a static gap at a fixed rate, determined according to the flow conventions and national models in force. It aims to measure the impact on the value of the entity's own funds of an instantaneous shock of market rates of 200 bps on this gap at a fixed rate;

This indicator is subject to a regulatory limit (20%);

The calculation of the discounting of this fixed rate gap varies from one institution to another;

It can be calculated in several ways:

• Either an NPV calculation;
• Or a simplified calculation.

This indicator has the advantage of being simple to implement by relying on the static gap at a fixed rate and does not necessarily require the recalculation of an NPV for the -200 and +200 scenario. This indicator also makes it possible to measure the sensitivity to a lower rate shock by a simple rule of three but does not meet the latest IRRBB regulatory requirements on the OT.

It is therefore necessary to complete the Basel 2 indicator with a calculation of the sensitivity of the NPV by means of a complete calculation of the cash flows and by taking into account the implicit and explicit options.

This NPV calculation can be based on an amortized cost or on a market valuation. This is where the EVE comes in, which will be introduced into the OT calculation in accordance with the latest regulatory requirements.

Scope of the interest rate risk - Focus on the EVE :

EVE (Economic Value of Equity)

Banks are required to comply with a list of constraints imposed by the regulations, in particular concerning the main regulatory ratio (Outlier Test or OT) to which we will introduce a notion of EVE.

• EVE (Economic Value of Equity) represents the valuation of balance sheet and off-balance sheet products and is generally calculated by discounting future cash flows. EVE is calculated to measure the impact of interest rates on the price of on- and off-balance sheet instruments. It reflects the evolution of product prices and their sensitivity to interest rates.
• Banks must ensure that the maximum variation of the EVE and thus the differential in the valuation of banking products compared to a reference scenario is controlled and de facto brought back to a regulatory capital threshold.
• The control of EVE, in this sense, is added to the monitoring of the Net Interest Margin (no regulatory constraints on this last ratio)
• EVE modelling: The EVE technique is based on the quantification of changes in the Net Present Value (NPV) of the various balance sheet and off-balance sheet items.

      This method quantifies the calculation of the risk of loss of the EVE of the Own Funds according to the different scenarios.

In summary:

1. Scenarios on currencies and interest rates are established on bank data;
2. Discounting of cash flows (out-of-scope option risk calculation);
3. Calculation of EV;
4. Quantification of the EV loss measure.

Interest Rate Risk - IRRBB Regulatory Constraints :

Regulatory requirements and constraints of banking institutions (for impacts on EBI)

• Disclosure of IRRBB exposure levels including ∆EVE and ∆MNI (Net Interest Margin) changes
• Transactions including the ∆EVE calculation should apply to a run-off balance sheet (balance sheet items are amortised and are not replaced by new transactions)
• Equity should be excluded from the calculation of exposures that contain all interest rate sensitive on/off balance sheet transactions
• Banking institutions are required to compare the calculation of maximum ∆EVE over the six scenarios (flattening curve, rising short rates, falling short rates, steepening.) with 15% of the value of equity (T1 according to the Basler standard).

Regulatory IRRBB (for impacts on EBI)

• Measurement of IRRBB: Institutions should measure the interest rate risk exposure of their banking book, both in terms of the possible evolution of the economic value (EV) and the evolution of the net interest margin (NIM) or the expected results;
• Interest rate shock scenarios: institutions should regularly measure the sensitivity of EV and NIM/earnings under different scenarios of changes in the level and shape of the interest rate yield curve and changes in the relationship between different market rates (i.e. basis risk);
• Internal governance arrangements: institutions should implement robust internal governance arrangements with respect to IRRBB;
• IRRBB policies: institutions should have properly justified, robust and documented policies to deal with all IRRBB issues of importance to their particular circumstances.
• Prudential standard shock: institutions should report to the competent authority the change in economic value resulting from the calculation of the outcome of the standard shock, as provided for in Article 98(5) of Directive 2013/36/EU and in this guidance;
• Internal capital: institutions should demonstrate that their internal capital is commensurate with the level of interest rate risk in their banking book under Pillar 2 under options combining the IRB and EVE measures

Interest rate risk - Other regulatory constraints related to interest rate risk and market risk :

INTERNAL CAPITAL REQUIREMENTS

• Internal capital adequacy assessment process: Pillar 2 of BIII strengthens the supervisory system and establishes the principles of a structured dialogue with the ACPR on the control of risks incurred by the institution and the adequacy of capital in relation to these risks. This dialogue must be based on four areas: mapping work, governance review, the internal capital adequacy assessment process (ICAAP) and the performance of stress tests;
• The latest EBA and BCBS IRRBB consultations have strengthened Pillar 2 of BIII for interest rate risk: institutions must now demonstrate that their internal capital is proportional to the level of interest rate risk in their banking book under Pillar 2 according to options combining the IRB and EVE measures (market practices favour an MNI sensitivity approach).

DECREE OF NOVEMBER 3, 2014 ON INTERNAL CONTROL

• RACI on internal control (report drawn up pursuant to Articles 258 to 266 and submitted annually to the ACPR): which details, among other things, the system for measuring and monitoring interest rate risk (and methodology);
• Measurement of Market Risk (Articles 122 to 133): relating to the identification, measurement and control of Market Risk;
• The measurement of Global Interest Rate Risk (Articles 134 to 139): relating to the identification, measurement and control of Interest Rate Risk.

NEW REGULATORY FRAMEWORK FOR MARKET RISK MONITORING (FRTB)

This reform aims to:

• Strictly defining the boundary between the Banking Book and the Trading Book;
• Replace the Value At Risk by the Expected Shortfall;
• Ensuring better management of credit risk.

Credit Market Liquidity risk Modeling and Machine Learning Deep Learning

Credit Risk Modeling and Machine Learning / Deep Learning

Credit risk modeling has undergone a major transformation with the advent of machine learning, deep learning, artificial intelligence, and data science. Here is how these technologies have specifically impacted credit risk modeling:

• More sophisticated models:

Non-linear Modeling: Machine learning and deep learning models can capture complex non-linear relationships between variables, unlike some traditional models.

Neural Networks: These structures allow for deep learning capabilities, capturing patterns in data that traditional methods might miss.

• Use of new data sources:

Advanced data science techniques allow for the incorporation of unstructured data (like texts, images, web browsing logs) into modeling, which wasn’t commonly done with traditional methods.

• Better validation and backtesting

Thanks to more advanced techniques, it's possible to conduct more robust cross-validation and backtesting of models to ensure they are both solid and predictive.

• Bias reduction and improved fairness

Some machine learning tools are specifically designed to detect and minimize biases in predictive models, which can help make credit risk modeling more fair.

• Automation and scalability:

AI-based models can be automatically updated based on new data, allowing for better adaptability in the face of market changes.

• Explainability and interpretability

While many advanced machine learning models (like deep neural networks) are viewed as "black boxes", there are now dedicated methods and tools to improve their interpretability.

Techniques:

Techniques such as boosting, bagging, or stacking combine predictions from multiple models to enhance accuracy and robustness of credit risk modeling.

Challenges:

• Complexity: Implementing deep learning models or advanced techniques requires deep expertise and greater computational resources.
• Overfitting: Overly complex models might "memorize" the training data instead of generalizing from it, making them less performant on new data.
• Regulation and compliance: In many jurisdictions, financial institutions must be able to explain how they assess credit risk. This can be a challenge with some machine learning or deep learning models.
• Data dependency: Machine learning models are highly dependent on data quality. If the training data is not representative or contains errors, models might produce inaccurate predictions.

EXAMPLES

Concrete examples of the application of machine learning and deep learning to credit risk modeling:

• Credit Scoring with Random Forest:

Description: A lender wants to improve its traditional logistic regression-based scoring model. To do this, they use Random Forest, an ensemble model based on multiple decision trees.

Data: Credit history, income, age, job, bank transactions, etc.

Results: The Random Forest, with its ability to capture non-linear relationships and interactions between variables, outperforms the traditional model in accuracy.

• Default Prediction with Deep Neural Networks:

Description: A financial institution wants to anticipate payment defaults on personal loans using a deep neural network.

Data: Payment history, credit usage, transaction data, demographic information, etc.

Results: DNNs manage to model complex relationships and incorporate information from a wide variety of sources, offering better performance than conventional methods.

• Social Media Text Analysis for Risk Assessment:

Description: Models of natural language processing (NLP) can be used to assess credit risk by analyzing borrowers' profiles on social networks.

Data: Posts, user interactions and behaviors on social media.

Results: NLP-based models, like LSTM or Transformers, can extract pertinent information about borrower behaviors, thus contributing to better risk assessment.

• Using Image Data for Credit Evaluation:

Description: A company uses image data, such as photos of owned assets or places of residence, to assess living standards and adjust credit scoring.

Data: Images uploaded or shared by the borrower.

Results: CNNs (Convolutional Neural Networks) process these images to extract useful features which, combined with other data, enhance risk assessment.

Fraud Detection:

Description: A bank is looking to detect fraudulent credit applications using Gradient Boosting.

Data: Details of credit applications, transaction history, geolocation information, etc.

Results: Gradient Boosting, with its capacity to handle vast amounts of data and assign weights to misclassified observations, is effective in detecting anomalies and potential fraud cases.

The impact of machine learning, deep learning, AI, and data science on credit risk modeling is profound. These technologies provide new methods for building, validating, and deploying risk models.

Market Risk Modeling and Machine Learning / Deep Learning 

Market risk modeling has also experienced significant advancements with the integration of machine learning, deep learning, artificial intelligence, and data science. Below is an exploration of how these technologies have influenced market risk modeling:

Enhanced Model Sophistication:

• Non-linear Relationships: Machine learning and deep learning models excel at capturing intricate, non-linear relationships among market variables, offering advantages over traditional linear models.
• Deep Neural Networks: Deep learning enables the creation of neural networks with multiple hidden layers, allowing for the detection of complex patterns in market data.

Utilization of New Data Sources:

• Unstructured Data Integration: Advanced data science techniques empower market risk models to incorporate unstructured data sources, such as news articles, social media sentiment, and alternative data sets, which traditional models often overlook.

Improved Validation and Backtesting:

• Cross-validation and backtesting benefit from more advanced techniques, ensuring that market risk models are robust, accurate, and capable of forecasting market movements effectively.

Bias Reduction and Fairness Enhancement:

• Specialized machine learning tools can identify and mitigate biases within predictive models, promoting fairness and equity in market risk assessment.

Automation and Scalability:

• AI-based models can be automated to adapt to new market conditions and data, ensuring that risk assessments remain relevant and up-to-date.

Explainability and Interpretability:

• While deep learning models are often considered "black boxes," efforts are made to enhance their interpretability through various techniques and tools, making it easier to understand the reasoning behind their predictions.

Techniques:

Various techniques can be applied in market risk modeling with machine learning and deep learning:

• Ensemble Methods: Techniques like boosting, bagging, and stacking can combine predictions from multiple models, improving the accuracy and robustness of market risk assessments.

Challenges:

Implementing machine learning and deep learning in market risk modeling presents its own set of challenges:

• Complexity: Deep learning models can be complex and require significant computational resources, as well as specialized expertise for implementation.
• Overfitting: Overly complex models may overfit the training data, hindering their ability to generalize effectively to new market conditions.
• Regulation and Compliance: Financial institutions often face regulatory requirements to explain their risk assessment methods, which can be challenging with complex machine learning models.
• Data Dependency: The quality and representativeness of training data significantly impact the accuracy and reliability of machine learning-based market risk models.

EXAMPLES:

Here are concrete examples illustrating the application of machine learning and deep learning to market risk modeling:

Stock Price Forecasting with Recurrent Neural Networks (RNNs):

• Description: A financial firm uses RNNs to predict stock price movements based on historical market data, news articles, and social media sentiment analysis.
• Data: Historical stock prices, trading volumes, news articles, and sentiment scores.
• Results: RNNs capture temporal dependencies and subtle market trends, providing improved stock price forecasts compared to traditional models.

Volatility Prediction with Support Vector Machines (SVMs):

• Description: An investment bank employs SVMs to predict market volatility based on various macroeconomic indicators and geopolitical events.
• Data: Economic data, political events, historical volatility indices.
• Results: SVMs excel at modeling complex relationships between market factors, leading to more accurate volatility forecasts.

Portfolio Optimization with Reinforcement Learning:

• Description: An asset management company uses reinforcement learning algorithms to optimize investment portfolios based on risk-return trade-offs and market conditions.
• Data: Historical financial data, portfolio holdings, market indicators.
• Results: Reinforcement learning enables dynamic portfolio adjustments in response to changing market dynamics, potentially increasing returns while managing risk.

Sentiment Analysis for Event-Driven Trading:

• Description: Hedge funds utilize sentiment analysis models to identify trading opportunities triggered by significant market-moving events.
• Data: News articles, social media posts, earnings reports.
• Results: Sentiment analysis models, often based on natural language processing techniques, provide valuable insights into market sentiment and potential price movements.

Machine learning, deep learning, AI, and data science have revolutionized market risk modeling, offering novel approaches to building, validating, and deploying risk assessment models in today's complex financial landscapes. These technologies are vital tools for managing and mitigating market risk effectively.

Liquidity Risk Modeling and Machine Learning / Deep Learning

Liquidity risk modeling, a critical aspect of financial risk management, has also evolved significantly with the integration of machine learning, deep learning, artificial intelligence, and data science. Here's an exploration of how these technologies have reshaped liquidity risk modeling:

Advanced Model Sophistication:

• Non-linear Relationships: Machine learning and deep learning models can capture intricate non-linear relationships among various liquidity indicators, which traditional models often struggle to address.
• Deep Neural Networks: Deep learning architectures enable the creation of neural networks with multiple hidden layers, facilitating the identification of complex liquidity patterns and anomalies.

Utilization of New Data Sources:

• Incorporation of Unstructured Data: Advanced data science techniques allow liquidity risk models to include unstructured data sources like news articles, social media sentiment, and transaction data, providing a more comprehensive view of liquidity dynamics.

Improved Validation and Stress Testing:

• Machine learning and deep learning techniques enhance the robustness of liquidity risk models, enabling more rigorous validation and stress testing to assess their resilience under various market conditions.

Bias Reduction and Fairness Enhancement:

• Specialized machine learning tools can identify and mitigate biases within liquidity risk models, promoting fairness and equity in liquidity assessment.

Automation and Scalability:

• AI-based models can be automated to adapt to changing market dynamics, ensuring that liquidity risk assessments remain relevant and responsive to real-time conditions.

Explainability and Interpretability:

• Despite their reputation as "black boxes," efforts are made to enhance the interpretability of deep learning models in liquidity risk modeling, making it easier to understand the rationale behind their predictions.

Techniques:

Various techniques can be applied in liquidity risk modeling with machine learning and deep learning:

• Time Series Analysis: Machine learning models can effectively analyze time series data to forecast liquidity trends and identify potential liquidity shocks.
• Anomaly Detection: Deep learning models excel at identifying anomalies in liquidity patterns, helping to detect liquidity crises or unusual market conditions.

Challenges:

Implementing machine learning and deep learning in liquidity risk modeling presents specific challenges:

• Complexity: Deep learning models can be complex and resource-intensive, demanding specialized expertise and computational resources.
• Overfitting: Overly complex models may overfit the training data, hindering their ability to generalize accurately to new liquidity scenarios.
• Regulation and Compliance: Financial institutions often need to explain their liquidity risk assessment methods to regulators, which can be challenging with complex machine learning models.
• Data Dependency: The quality and representativeness of training data significantly impact the accuracy and reliability of machine learning-based liquidity risk models.

EXAMPLES:

Here are concrete examples illustrating the application of machine learning and deep learning to liquidity risk modeling:

Liquidity Stress Testing with Recurrent Neural Networks (RNNs):

Description: A bank uses RNNs to conduct liquidity stress tests by analyzing historical liquidity data and assessing the impact of various market scenarios.

• Data: Historical liquidity metrics, market events, macroeconomic indicators.
• Results: RNNs capture temporal dependencies and complex liquidity dynamics, offering more accurate stress test results compared to traditional methods.

Liquidity Risk Early Warning System with Random Forest:

Description: A financial institution develops an early warning system using Random Forest, which leverages various liquidity indicators and alternative data sources to predict liquidity shortages.

• Data: Liquidity metrics, news sentiment, transaction data.
• Results: Random Forest models provide timely alerts for potential liquidity issues, enabling proactive risk management.

Liquidity Forecasting with LSTM (Long Short-Term Memory) Networks:

Description: A hedge fund employs LSTM networks to forecast liquidity needs for its trading strategies based on historical transaction data and market volatility.

• Data: Transaction history, market data, trading volumes.
• Results: LSTM networks capture sequential dependencies in transaction data, facilitating accurate liquidity forecasts for optimizing trading strategies.

Liquidity Risk Monitoring through Social Media Sentiment Analysis:

Description: An investment firm incorporates sentiment analysis of social media conversations and news articles to monitor liquidity risk associated with specific assets or markets.

• Data: Social media content, news articles, asset-specific indicators.
• Results: Sentiment analysis models contribute valuable insights into market sentiment, helping assess liquidity risk more comprehensively.

Machine learning, deep learning, AI, and data science have revolutionized liquidity risk modeling, providing innovative approaches to building, validating, and deploying liquidity assessment models in today's complex financial landscape. These technologies are indispensable tools for effectively managing and mitigating liquidity risk.

We work in collaboration with several fintechs

The Fintechsa aim to replace certain activities or even the profession of banker. For now, they are mainly positioned in niche markets but are tending to develop in order to offer customers and companies the equivalent of the services they can obtain in a bank.

The public has become very interested in the solutions proposed by Fintech, because they make it possible to avoid using banks as intermediaries. Fintech companies make services (previously opaque) only offered to the wealthiest customers by traditional banks, accessible to everyone and at low cost.

These societies are disruptive, that is, they break established habits by offering a new way of consuming services that is more accessible and less expensive. Finally, they also contribute to the elimination of current players by proposing a new business model.

Banks are closely following the Fintechs because they know that the banking profession will be redefined in the coming years with the arrival of many innovative technologies

Big data, artificial intelligence and blockchain are the main technological revolutions that will revolutionize the banking sector. Unfortunately, much remains unknown. Indeed, we do not yet know if the blockchain will be viable, for example.  But unlike the blockchain, big data and artificial intelligence should certainly change the habits of the financial sector....

The transformation of your traditional activities

The advent of Fintech does not necessarily signify the decline of traditional players. On the contrary, this evolution can be seen as an opportunity. As a traditional player, it is essential to grasp how and to what extent Fintech affects your industry.

How are Fintechs revolutionizing the financial services market? How can one capitalize on these transformations and the innovations they produce? These questions resonate with you, and we are here to answer them!

For each of your business segments, we offer to assess your current position and conduct an impact study based on two criteria:

• Magnitude: measuring the depth of the changes that traditional players need to make in order to adapt and benefit from the rise of Fintech. This scale ranges from 1 to 10: the higher the score, the deeper the change.

• Timeframe: reflecting the number of years before these adjustments become imperative. Also rated from 1 to 10, a higher score indicates an urgency to act swiftly.

Our commitment is to provide you with lasting and technologically advanced solutions.

• Collaboration Over Confrontation: The rise of Fintech is not necessarily a threat. Collaborations and partnerships can be established, combining the innovative strength of Fintech with the expertise and reliability of traditional players.

• Digital Adaptability: With the increasing digital expectations of customers, adapting to Fintech technology can enhance the customer experience, increase operational efficiency, and unveil new revenue streams.

• Regulation and Compliance: Traditional players have the advantage of in-depth knowledge of financial sector regulations. Integrating this expertise with Fintech solutions can offer services that are both innovative and compliant.

• Culture of Innovation: Embracing Fintech can lead to a stronger innovation culture within traditional organizations, making them more competitive and resilient to future changes.

In summary, as the financial landscape rapidly evolves, we believe the key lies in collaboration, innovation, and adaptability. The merger of the strengths of traditional players and Fintech can forge a more robust and inclusive financial ecosystem for all.