General Information & Client Services
  • Americas: +1.212.553.1653
  • Asia: +852.3551.3077
  • China: +86.10.6319.6580
  • EMEA: +44.20.7772.5454
  • Japan: +81.3.5408.4100
Media Relations
  • New York: +1.212.553.0376
  • London: +44.20.7772.5456
  • Hong Kong: +852.3758.1350
  • Tokyo: +813.5408.4110
  • Sydney: +61.2.9270.8141
  • Mexico City: +001.888.779.5833
  • Buenos Aires: +0800.666.3506
  • São Paulo: +0800.891.2518

This article discusses how to address the specific challenges that IFRS 9 poses for retail portfolios, including incorporating forward-looking information into impairment models, recognizing significant increases in credit risks, and determining the length of an instrument's lifetime. We describe two approaches to analyzing retail portfolios, suggest practical interpretations of IFRS 9 guidelines, and answer common questions pertaining to retail portfolios.

Introduction

When building and implementing econometric models for different asset classes, the modeler needs to carefully examine the requirements from the perspective of the final users of the models. A trader of whole loans may be more interested in the accurate modeling of loan-level cash flows and exploiting any statistical arbitrage. A servicer is likely to be concerned about delinquency transitions and time to liquidation. Regulatory stress testing requires that the models demonstrate sensitivity to macroeconomic conditions. Risk management requires that the models correctly capture the correlation between different assets in the portfolio.

The IFRS 9 guidelines pose some interesting challenges, including the following:

  • An important consideration in the impairment model in IFRS 9 is the use of forward-looking information in the models. Decisions around classification of assets into different stages and the calculation of the expected credit losses require consideration of forward-looking macroeconomic information.
  • A classification of assets into different stages requires determination of a significant increase in the credit risk.
  • Unbiased point-in-time estimates of the expected credit losses have to be computed by using a probability-weighted amount that is determined by evaluating a range of possible outcomes.
  • Depending on the stage into which an asset is classified, either a 12-month or a lifetime expected credit loss may have to be determined.

This paper addresses these considerations for retail portfolios. Retail portfolios can be analyzed with either a bottom-up approach or a top-down one. We discuss each of these approaches in more detail below.

Bottom-Up Approach

A bottom-up approach involves constructing loan-level models for each loan in the portfolio. Results can be aggregated over all the loans in different cohorts or segments to arrive at segment-level or portfolio-level results. Loan-level models are usually hazard rate models and can be constructed in a competing risk framework. The data is naturally organized as panel data; each loan has multiple observations through time. Defaults and prepayments compete with each other in a multi-period setting. Survival models in this framework can be built using a panel logit model.

Loan-level models are usually hazard rate models and can be constructed in a competing risk framework. The data is naturally organized as panel data; each loan has multiple observations through time. Defaults and prepayments compete with each other in a multi-period setting. Survival models in this framework can be built using a panel logit model.

A bottom-up approach has the advantage that the results are naturally available at the highest level of granularity. Heterogeneity of the loan characteristics can be easily accommodated.

A bottom-up approach has the advantage that the results are naturally available at the highest level of granularity. The explanatory variables, such as loan and borrower characteristics and macroeconomic variables, are used at the loan level. Likewise, the performance variables, such as defaults, prepayments, cash flows, and losses, are modeled at the loan level. Heterogeneity of the loan characteristics can be easily accommodated.

Building loan-level models requires reliable historical loan-level data. This can be onerous and expensive. If the loan-level data is not reliable, the models that are built may have to be recalibrated. The implementation can also require additional resources. In situations where the portfolio consists of a large number of homogeneous assets, a loan-level approach may not be necessary.

Top-Down Apprach

A top-down approach involves segmenting the portfolio by vintage and risk characteristics. The segments can be as coarse or as granular as required. For example, a large homogeneous portfolio of retail credit cards could be modeled as a single segment. On the other hand, a portfolio may be segmented by vintage, geographic regions, or the borrower's risk profile. Different segments can have different models. Results can be either aggregated further to arrive at the portfolio-level results or calibrated further to instrument-level figures.

Segment-level models are easier to build because they typically require less data. The performance variables of interest can be directly modeled as a function of segment-level characteristics. Models can be implemented faster.

Segment-level models are better suited to homogeneous portfolios. When a portfolio consists of heterogeneous assets, several segments are needed in order to accurately model the portfolio along multiple dimensions. If portfolio composition changes through time or if assets migrate from one segment to another, greater care is needed in segmenting the portfolio.

The suitability of a particular model will thus depend on the type of collateral and the portfolio composition.

Next, we turn our attention to specific guidelines in the IFRS 9 standard, how they apply to retail portfolios, and how we can address them while building and implementing the models.

How to Use Forward-Looking Information

One of the key issues identified with IAS 39 impairment regulation is that only past events and current conditions can be considered in measuring credit losses. This leads to a notable weakness in the models developed under IAS 39 standards: that there can be delayed recognition of credit losses. In addition to considering past events and current conditions, the new standard requires that forecast information must be used in measuring Expected Credit Losses (ECL) if available without undue cost or effort. Forward-looking information is to be used for stage allocation as well as for the calculation of the ECL. We discuss how forward-looking information can be incorporated in the models. The use of this information for stage allocation and for calculation of the ECL is discussed in later sections.

An econometric model of the retail assets, whether it is done at the cohort level or the loan level, involves relating the performance of the assets to macroeconomic factors. Once this relationship is established, forecasting the losses or determining the lifetime PD requires using these models on prescribed scenarios.

One way to include forward-looking information is to incorporate econometric panel data models that will give risk parameter forecasts under multiple scenarios. As the stage allocation should use the change in the risk of a default occurring over the expected life of the financial instrument, banks need to determine the extent to which forward-looking information will be included in lifetime PDs.

Our recommendation is to incorporate forward-looking information into the assessment of lifetime PDs for lending originated solely after the implementation of IFRS 9.

Our interpretation of the quantitative metric required for determining stage allocation is the change in the lifetime PD of an instrument since origination, relative to age. The caveat that the change is relative to age is essential for retail portfolios, as the lifetime PD will depend on time until derecognition (retail portfolios show a strong lifecycle component: nonlinear relationship between PD and time-since-origination). This is highlighted in appendix paragraph B5.5.11 of the standard:

Because of the relationship between the expected life and the risk of a default occurring, the change in credit risk cannot be assessed simply by comparing the change in the absolute risk of a default occurring over time. For example, if the risk of a default occurring for a financial instrument with an expected life of 10 years at initial recognition is identical to the risk of a default occurring on that financial instrument when its expected life in a subsequent period is only five years, that may indicate an increase in credit risk. This is because the risk of a default occurring over the expected life usually decreases as time passes if the credit risk is unchanged and the financial instrument is closer to maturity.

An econometric panel data modeling approach can help identify a forward-looking lifetime PD at the latest reporting date, but a question is raised concerning if and how to determine a forward-looking lifetime PD at origination. To use forward-looking information historically, one could either leverage historical macroeconomic scenarios on a monthly basis, or adjust origination PDs with historical macroeconomic data (i.e., utilizing what has actually happened in the economy since origination).

Our recommendation is to incorporate forward-looking information into the assessment of lifetime PDs for lending originated solely after the implementation of IFRS 9. For lending originated prior to the implementation of IFRS 9, lifetime PDs at origination can reflect the assessment of credit risk at the time of origination, which may not include forward-looking information. The usage of historical origination PDs for instruments originated prior to IFRS 9 implementation is justified by the following standards:

  • Prior to the introduction of IFRS 9, there was no explicit requirement for forward-looking information to be used to adjust historical estimates of PD.
  • The work required to adjust historical PDs to incorporate forward-looking information would be considerable, going against the clause to "use information that is available without undue cost or effort" (B7.2.2).

The rationale behind stage allocation, which requires origination PD, is to compare the current view of default risk with the view that was held when the lending was agreed and the product was priced. To adjust the origination PDs for forward-looking information (or any additional data) would be inconsistent with this aim. The only adjustments that should be made to origination PDs are ensuring these are unbiased point-in-time best estimates of the lifetime PD.

In summary, our recommendation to address the forward-looking aspect of the standard is to use panel data (vintage or loan-level models) using macroeconomic drivers for retail portfolios. These granular-level outputs can be calibrated to instrument-level figures, if required, before calculating instrument-level IFRS 9 impairment. The inclusion of macroeconomic variables allows the estimation of ECL under several different scenarios and the generation of probability-weighted outcomes. This approach captures both a range of forecasts and the non-linearity in the ECL calculation.

How to Calculate Unbiased Point-in-Time Estimates

Paragraph 5.5.17(a) of the standard states that an entity shall measure ECL of a financial instrument in a way that reflects an unbiased amount. Therefore, banks need to consider if and how their existing capital models and methodologies can be leveraged. This decision will likely be driven by the level at which downturn adjustments are incorporated into their existing models.

Banks that lack suitable models for IFRS 9 purposes can use a panel data modeling approach where the data is split by vintages. For PD modeling, the vintages refer to the month or quarter of origination, whereas for LGD modeling, the vintages refer to the month or quarter of default. Data can be split by further levels of client-specific segmentation such as product type, region, or LTV band.

Factors influencing vintage-segment performance can be conceptually divided into three classifications: the lifecycle trends depending on a loan's age-on-books (seasoning), the factors indicating the quality of a vintage, and the characteristics of the current economic environment that depend only on calendar time. Other effects also operate across more than one of these main categories and can be modeled as interactions between them. An examination of each of these types of effects in isolation is essential to understanding the multidimensional nature of the data and the models used to forecast it. Using this approach on a panel data of marginal default rates and loss rates can help provide 12-month and lifetime ECL.

Consider a bank that already has 12-month Point-in-Time (PiT) Basel models or 12-month Through-the-Cycle (TTC) models with an easily extractable PiT component. The bank can achieve IFRS 9 compliance through a scaling process that leverages the vintage-level outputs to provide account-level lifetime expected credit losses that are consistent with the Basel 12-month PiT outputs.

For banks that use 12-month TTC models, with no possibility of extracting PiT outputs, our proposal is to use a vintage-level panel data modeling approach to estimate 12-month and lifetime ECL. The distribution of TTC PDs for a given vintage can then be used as a benchmark to define the ranking distribution of the account PiT PDs around the new vintage-level mean.

When using a loan-level model, an unbiased point-in-time estimate can be obtained by using models that incorporate lifecycle effects and macroeconomic factors. Using survival models, the dependence of the PD of a loan on its seasoning can be captured. Therefore, newly originated loans will behave differently from seasoned loans. Loan-level models use past information through changes in macroeconomic variables such as home prices and unemployment rates from loan origination to the reporting date, and forward-looking information such as future changes in macroeconomic conditions. The output of the loan-level models is a conditional PD or LGD. Through the use of probability-weighted scenarios described later in the paper, a point-in-time estimate of the ECL can be obtained.

How to Define a Lifetime View of ECL

Paragraph 5.5.4 of the standard states:

The objective of the impairment requirements is to recognise lifetime expected credit losses for all financial instruments for which there have been significant increases in credit risk since initial recognition.

We address two challenges relating to providing a lifetime view of risk parameters: first, how to determine the length of an instrument's lifetime, and second, how to model risk parameters over the lifetime.

In regards to determining the length of an instrument's lifetime, the standard states in paragraph 5.5.19:

The maximum period to consider when measuring expected credit losses is the maximum contractual period (including extension options) over which the entity is exposed to credit risk and not a longer period, even if that longer period is consistent with business practice.

With retail products in mind, this raises the following two questions for how to define the lifetime:

  1. When the terms and conditions of an instrument are amended, should this lead to the derecognition of the financial asset and the recognition of a new asset?
  2. For revolving products such as credit cards or overdraft facilities where the contractual period can be as little as one day, should the lifetime for these products only be one day?
Lifetime Definition – Question 1: Derecognizing Assets

Chapter 3 of the standard answers the first question by defining both recognition and derecognition. Derecognition occurs through both the expiration of the contractual period and financial asset transfer. This asset transfer can be identified by the transferral of the contractual rights to receive the cash flows of the financial asset. There is also a case for asset transfer where the rights are retained, as detailed in paragraph 3.2.5 of the standard:

When an entity retains the contractual rights to receive the cash flows of a financial asset (the ‘original asset'), but assumes a contractual obligation to pay those cash flows to one or more entities (the ‘eventual recipients'), the entity treats the transaction as a transfer of a financial asset if, and only if, all of the following three conditions are met:
(a) The entity has no obligation to pay amounts to the eventual recipients unless it collects equivalent amounts from the original asset. Short-term advances by the entity with the right of full recovery of the amount lent plus accrued interest at market rates do not violate this condition.
(b) The entity is prohibited by the terms of the transfer contract from selling or pledging the original asset other than as security to the eventual recipients for the obligation to pay them cash flows.
(c) The entity has an obligation to remit any cash flows it collects on behalf of the eventual recipients without material delay. In addition, the entity is not entitled to reinvest such cash flows, except for investments in cash or cash equivalents (as defined in IAS 7 Statement of Cash Flows) during the short settlement period from the collection date to the date of required remittance to the eventual recipients, and interest earned on such investments is passed to the eventual recipients."
Lifetime Definition – Question 2: Challenges in Retail Revolving Credit

The second question is addressed in paragraph 5.5.20:

Some financial instruments include both a loan and an undrawn commitment component and the entity's contractual ability to demand repayment and cancel the undrawn commitment does not limit the entity's exposure to credit losses to the contractual notice period. For such financial instruments, and only those financial instruments, the entity shall measure expected credit losses over the period that the entity is exposed to credit risk and expected credit losses would not be mitigated by credit risk management actions, even if that period extends beyond the maximum contractual period.

In order to measure over the expected period that these entities are exposed to credit risk, we propose a collective assessment of lifetime length using behavioral data. A panel data modeling approach can be leveraged to model the proportion of a vintage's instruments that are recognized over age.

For non-revolving products, our interpretation for defining the lifetime view is more straightforward. The maturity date of the contractual period helps provide the end date for the calculation of each instrument's lifetime length. Thus, for any given reporting date, the remaining lifetime over which to determine risk parameters is simply the time to maturity.

Calculating Probability-Weighted Expected Credit Losses

The measurement of an expected credit loss requires calculation of expected present value of the cash shortfalls. These credit losses are to be weighted using the probability of default. Since the models for the PD and LGD use macroeconomic drivers and loan and borrower characteristics, the calculation of the expected credit losses involves projecting the PD, LGD, and cash flows for different macroeconomic scenarios. An entity need not consider every possible scenario. However, different scenarios with their probabilities of occurrence must be considered.1

Defining a Significant Increase in Credit Risk

Stage allocation requires determining if an asset has undergone a significant increase in credit risk since initial recognition. Paragraph 5.5.9 of the standard defines the significant increase in credit risk as a significant "change in the risk of a default occurring over the expected life of the financial instrument." This suggests that the decision should be based on the change in the lifetime PD since origination; however, there is little guidance around what quantifies a significant change. We need to first define this change in the risk of default and then set a threshold to determine what constitutes a significant increase. There are many options for the exact metric by which to allocate instruments into stages for retail portfolios.

Absolute Change

The absolute change in the lifetime PD since origination is the simplest metric for calculating the change, but would likely result in complexity for the threshold assessment. The age of the instrument at the latest reporting date would be a key driver, leading to an age-specific threshold.

Other dimensions that might warrant consideration are the length of the remaining lifetime and the size of the lifetime PD at origination. If the lifetime PD at origination is very low, the instrument could still be classified within stage 1 after an increase in risk if the PD is still considered low risk. This is documented in paragraph 5.5.10, which states, "An entity may assume that the credit risk on a financial instrument has not increased significantly since initial recognition if the financial instrument is determined to have low credit risk at the reporting date." The approach of basing the stage allocation off the absolute change in credit risk is not recommended as there are so many dimensions that would require assessment.

Relative Change

The relative change in the lifetime PD since origination would similarly require a number of dimensions in assessing the threshold. The assessment might be slightly simpler than using the absolute change, as the size of the lifetime PD at origination might not need to be a dimension. However, the age at latest reporting date and length of remaining lifetime would both need consideration, unless combined into a field to measure the percentage of expected lifetime remaining. This approach is not recommended due to the complexity of capturing the age differences between the lifetime PD at origination and lifetime PD at latest reporting date.

Absolute Change in Age-Specific Lifetime PD Forecast

The absolute change in the age-specific lifetime PD forecast compares the lifetime PD at the latest reporting month (at age α) with the lifetime PD forecast at origination for the instrument once it reaches age α. In Figure 1, a vintage of instruments are modeled using the panel data modeling approach to give a vintage lifetime PD curve. The panel data approach models marginal default rates that can be accumulated to give the lifetime PD curve. Similarly, the marginal default rates can be accumulated into 12-month PD curves over age. The vintage 12-month PD curve can be calibrated against Basel 12-month PDs to give instrument-specific 12-month PD curves through the remainder of the instrument's lifetime. Lifetime views of the Basel PDs can subsequently be extracted, with the lifetime PD curves calibrated to these points, as shown in the graph. The metric that can subsequently be used for setting the threshold is the absolute distance between Basel-calibrated lifetime PD curves at the age of the latest reporting date.

Figure 1. Lifetime PD
Lifetime PD
Source: Moody's Analytics
Relative Change in Age-Specific Lifetime PD Forecast

Whether the absolute distance differs across age would need to be assessed. It is likely that the age dimension would remain essential in setting the threshold using the absolute change. Our recommendation is thus to consider the relative change in the age-specific lifetime PD forecast. This has the caveat of paragraph 5.5.10 as mentioned, which states that low values at origination can have notable increases but still be classified as stage 1.

Considering the options for determining the metric above has helped identify some of the dimensions that need to be assessed when quantitatively setting a threshold for stage allocation. Below are some of the key areas for consideration:

  • The size of the lifetime PD at origination
  • The age of the instrument at the latest reporting date
  • The length of the lifetime remaining
  • The product type

Besides the quantitative assessment, entities also need to consider whether to rebut the presumption that a financial asset's PD has increased significantly since initial recognition when contractual payments are more than 30 days past due. On top of this, a qualitative assessment is recommended for identifying any changes in behavior that are not immediately captured in an entity's default definition.

Concluding Remarks

In this paper, we have addressed several important considerations in the modeling and implementation of the IFRS 9 standard for retail portfolios. We have shown how these guidelines should be interpreted and how they can be incorporated into loan-level and segment/ vintage-level models. We have looked to address some of the key IFRS 9 issues facing banks with a focus on a retail perspective. What is clear from discussing these issues with peers across the globe is that the standard leaves room for interpretation in defining methodologies to reach IFRS 9-compliant impairment models. Time will tell as to whether any specificities of methodology design become enforced by regulators.

Footnote

1 Further discussion on generating scenarios and their associated probabilities can be found in Black, Levine, & Licari, Probability-Weighted Outcomes Under IFRS 9: A Macroeconomic Approach, Risk Perspectives Magazine, June 2016, Moody's Analytics.

SUBJECT MATTER EXPERTS
As Published In:
Related Insights

CECL Quantification: Retail Portfolios

In this webinar, our experts discuss the important considerations in the modeling and implementation of the CECL standard for retail portfolios. Learn more about loan-level modeling approaches that can be used to forecast credit losses for retail portfolios and how to leverage existing risk measurement practices.

CECL Quantification: Retail Portfolios Webinar Slides

In this presentation from our webinar, our experts discuss the important considerations in the modeling and implementation of the CECL standard for retail portfolios. Learn more about loan-level modeling approaches that can be used to forecast credit losses for retail portfolios and how to leverage existing risk measurement practices.

CECL Quantification: Retail Portfolios Webinar Slides

In this webinar, our experts discuss the important considerations in the modeling and implementation of the CECL standard for retail portfolios. Learn more about loan-level modeling approaches that can be used to forecast credit losses for retail portfolios and how to leverage existing risk measurement practices.

U.K. Residential Mortgages Risk Weights: PRA Consultation Paper CP29/16

This paper presents best practices for addressing PRA Consultation Paper CP29/16.

October 2016 Pdf Dr. Juan M. Licari, Dimitrios Papanastasiou, Maria Valle del Olmo

Complying with IFRS 9 Impairment Calculations for Retail Portfolios

Gain insight in to how to address IFRS 9 challenges by incorporating forward-looking information into impairment models for consumer lending portfolios.

August 2016 WebPage Barnaby Black

Probablility-Weighted Outcomes Under A Macroeconomic Approach

Learn how to develop a framework that addresses the forward-looking and probability-weighted aspects of IFRS 9 impairment calculation using macroeconomic forecasts and scenarios.

August 2016 WebPage Barnaby Black

Probability-Weighted Outcomes Under IFRS 9: A Macroeconomic Approach

In this article, we discuss development of a framework that addresses the forward-looking and probability-weighted aspects of IFRS 9 impairment calculation using macroeconomic forecasts. In it, we address questions around the practical use of alternative scenarios and their probabilities.

June 2016 WebPage Barnaby Black, Glenn LevineDr. Juan M. Licari

Advanced Estimation and Simulation Methods for Retail Credit Portfolios: Frequentist vs. Bayesian Techniques

In this article, we compare the results of estimating retail portfolio risk parameters (e.g., PDs, EADs, LGDs) and simulating portfolio default losses using traditional – frequentist – methods versus Bayesian techniques.

December 2015 WebPage Dr. Juan M. Licari, Dr. Gustavo Ordóñez-Sanz, Chiara Ventura

Multi-Period Credit Risk Analysis: A Macro-Scenario Approach Presentation Slides

In this presentation, Dr. Juan Licari of Moody's Analytics will present an innovative framework for stochastic scenario generation that allows risk managers and economists to build multi-period environments, integrating conditional credit and market risk modeling to meet dynamic stress testing needs.

December 2015 Pdf Dr. Juan M. Licari

Market Risk Stress Testing Models Presentation Slides

In this presentation, Dr. Juan Licari presents a two-stage process that generates consistent, transparent scenario-specific forecasts for all relevant market and credit risk instruments, ensuring cross-consistency between projections for macroeconomic and financial series.

December 2015 Pdf Dr. Juan M. Licari

Market Risk Stress Testing Models

In this presentation we present a two-stage process that generates consistent, transparent scenario-specific forecasts for all relevant market and credit risk instruments, ensuring cross-consistency between projections for macroeconomic and financial series.

December 2015 WebPage Dr. Juan M. Licari

Multi-Period Credit Risk Analysis: A Macro-Scenario Approach

In this presentation, we present an innovative framework for stochastic scenario generation that allows risk managers and economists to build multi-period environments, integrating conditional credit and market risk modeling to meet dynamic stress testing needs.

November 2015 WebPage Dr. Juan M. Licari

IFRS 9 Impairment Webinar Series – Models for Implementation

This webinar discusses determining the best approaches for model development and governance for IFRS 9 Impairment calculations.

September 2015 WebPage Manuele Iorio, Dr. Juan M. Licari

Moody's Analytics Forecasts of RCA Commercial Property Price Indexes

This article discusses the methodology used in generating forecasts of the RCA commercial Property Prices Indices.

September 2015 Pdf Dr. Juan M. LicariOlga Loiseau-Aslanidi, Dr. José Suárez-Lledó

Multi-Period Stochastic Scenario Generation

Robust models are currently being developed worldwide to meet the demands of dynamic stress testing. This article describes how to build consistent projections for standard credit risk metrics and mark-to-market parameters simultaneously within a single, unified environment.

May 2015 WebPage Dr. Juan M. Licari, Dr. Gustavo Ordóñez-Sanz

Integrating Macroeconomic Scenarios into a Stress Testing Framework

This article describes the three principles that need to be understood and analyzed for banks to have a realistic chance of integrating alternative scenario work into their stress testing workflow.

November 2014 WebPage Dr. Juan M. Licari

Arbitrage-Free Scenarios for Solvency II

This article discusses a macroeconomic forecasting model that is able to generate arbitrage-free scenarios.

May 2014 WebPage Dr. Juan M. Licari, Dr. José Suárez-Lledó

Handling low default portfolios under stress

Regulators are challenging how to perform stress testing on low default portfolios by reviewing bank's PD models for RWA stress testing, in the absence of data they need to be convinced of the methodology used. In this Moody's Analytics webinar we put forward a statistical approach to stress testing low default portfolios with practical case studies

February 2014 Pdf Manuele Iorio, Dr. Juan M. Licari

Stress Testing for Retail Credit Portfolios: A Bottom-Up Approach

This article focuses on model building from a bottom-up perspective of mortgages and home equity lines of credit to underscore the importance of loan-level analytics.

November 2013 WebPage Dr. Shirish Chinchalkar

Gauging the Risk of Europe's Banks: What Might the ECB Find?

The European Central (ECB) has begun a year-long comprehensive assessment of the Euro area banking system. In this webinar, Moody's Analytics seeks to provide a default data-driven context for the ECB's exercise and a preview for what is to come.

November 2013 WebPage Danielle Ferry, Dr. Juan M. Licari

Account Level Retail Modeling

Why are mortgages complicated? In this presentation, Moody's Analytics expert Dr. Shirish Chinchalkar shares the challenges to modeling a retail portfolio – and how to solve them.

October 2013 Pdf Dr. Shirish Chinchalkar

Modelling and Stressing the Interest Rates Swap Curve

We present a two-step modelling and stress testing framework for the term structure of interest rates swaps that generates sensible forecasts and stressed scenarios out of sample. Our methodology is able to replicate two important features of the data: the dynamics of the spread across maturities and the alignment of the key swap rates tenor points to their corresponding government yields. Modern models of the term structure of interest rates typically fail to reproduce these and are not designed for stress testing purposes. We present results for the euro, the U.S. dollar, and British pound swap curves.

September 2013 Pdf Dr. Juan M. LicariOlga Loiseau-Aslanidi, Dr. José Suárez-Lledó

A Macroeconomic View of Stress Testing

This article discusses how developing deterministic scenarios form a macroeconomic view on stress testing that helps to uncover system or enterprise-wide vulnerabilities and assist banks in making more informed business decisions.

September 2013 WebPage Dr. Juan M. Licari, Dr. José Suárez-Lledó

Stress Testing of Retail Credit Portfolios

In this article, we divide the stress testing process for retail portfolios into four steps, highlighting key activities and providing details about how to implement each step.

September 2013 WebPage Dr. Juan M. Licari, Dr. José Suárez-Lledó

Moody's CreditCycle - Reverse Stress Testing Capabilities

Download this article to understand how Moody's CreditCycle can be used to carry out macroeconomic stress testing.

November 2012 Pdf Barnaby Black, Dr. José Suárez-Lledó

Reverse Stress Testing from a Macroeconomic Viewpoint: Quantitative Challenges & Solutions for its Practical Implementation

This whitepaper examines the challenge of multiplicity in reverse stress testing, where the same outcome can be obtained with multiple combinations of risk factors and economic scenarios.

October 2012 Pdf Dr. Juan M. Licari, Dr. José Suárez-Lledó

A Macro-finance View on Stress Testing

For most financial practitioners, stress-testing is a “must-do” activity, even if it is not a regulatory requirement. Such stress-testing encompasses a wide range of sophisticated and quantitative exercises, including assessments of market, credit and liquidity risks. This article discusses several approaches and outlines a foundation for a robust and consistent stress-testing framework.

May 2012 Pdf Andrea Appeddu, Dr. Juan M. Licari, Dr. José Suárez-Lledó

Reverse Stress Testing: Challenges and Benefits

Reverse stress testing is becoming recognised throughout the world for its benefits. This presentation explains what reverse stress testing is and what it can achieve, along with the challenges it presents. Here we show you why reverse stress testing can lead to a deeper understanding of an organisation's susceptibility to risk and why it is a valuable tool for any risk management strategy.

November 2010 Pdf Dr. Christian Thun, Dr. Juan M. Licari, Mark Zandi

Modeling and Stressing the Interest Rates Swap Curve

This article presents a two-step modeling and stress testing framework for the term structure of interest rates swaps that generates sensible forecasts and stressed scenarios out of sample. The results are shown for the euro, the US dollar, and British pound swap curves.

WebPage Dr. Juan M. LicariOlga Loiseau-Aslanidi, Dr. José Suárez-Lledó

Mortgage Models for CECL: A Bottom-Up Approach

In this article, we describe how a loan-level modeling approach can be used to forecast credit losses in residential mortgages, and review the challenges a bank may face in complying with the FASB’s recent Accounting Standards Update on reporting credit losses.