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: stochastic dynamic macro models. It gives a step-by-step breakdown of the development of a dynamic framework for stochastic scenario generation that allows risk managers and economists to build multi-period environments, integrating conditional credit and market risk modeling.
Author: Dr. Juan Licari & Dr. Gustavo Ordonez-Sanz
Date: June 5, 2015
Credit card charge‐off rates are still declining. According to CreditForecast.com, the U.S. average charge off rates over the last two quarters have been 3.2% annualized, down 0.28 percentage point on a year ago basis. Total outstanding balances during the same period increased 4.25% to $566 billion from a year earlier. Even taking into account the harsh winter, which weighed on consumer spending, both charge‐off rates and outstanding balances for credit cards look to be recovering.
Author: Jeesang Jung
Date: May 20, 2015
Many of the dynamic features of the auto finance industry are countercyclical. We argue that the CCAR auto modeling methodology used by the Fed pays too little heed to these features. We propose a plausible alternative scenario that synchronizes the observed dynamics to be maximally stressful for the auto finance sector.
Author: Juan Carlos Calcagno, Tony Hughes, Stephen Kernytsky
Date: June 24, 2013
The Comprehensive Capital Analysis and Review (CCAR) stress-testing protocol dictates that banks be able to quantify the effects of the Supervisory Stress Scenarios (SSS) on their entire profit and loss statement and balance sheet. Most banks that are subject to the requirements already have models in place for analyzing credit losses, though responses vary in terms of the quality of models and the level of coverage. This year, we feel that attention will increasingly turn to the problem of testing the other factors that can make or break a bank’s performance under stress. Notably, banks need a solution for analyzing the liabilities book while also extending analysis to cover the volume and quality of new originations made given the bank’s risk appetite and the revenues that flow from both new and existing loans on the asset side of the ledger.
Author: Tony Hughes
Date: June 13, 2013
Some of the pushback we’ve had from internal and external validators and the Fed during the last round of Comprehensive Capital Analysis and Review is that stress-testing models need to be very simple. Tony Hughes, senior director for consumer credit at Moody's Analytics, discusses that while it's important for bank managers to understand their stress testing models, the potential for these models to yield deep insights will be lost if we oversimplify them.
Author: Tony Hughes
Date: May 15, 2013
In this Banker & Tradesman article, John Baer talks about the empowering of deal makers through a thorough understanding the current risk, borrowing limits, exposure and collateral values for each counterparty within their customer's borrowing group, and across their portfolio.
Author: John Baer
Date: May 20, 2013
The main purpose of our exercise is to stress-test the elements of a standard, through-the-cycle rating transition matrix with an explicit and transparent connection to macroeconomic drivers. The challenge behind this exercise is evident in the time-series nature of through-the-cycle migrations: They are built to be stable over time, and changes happen in waves (of upgrades or downgrades) that disappear once the economic and credit conditions go back to normal. In other words, their behaviour over time is not symmetric around an average or median value. They actually show bimodal distributions, with observations accumulated in different states of nature: (i) normal and/or good times vs. (ii) stressed conditions.
Authors: Juan Licari, Jose Suarez-Lledo, Barnaby Black
Date: May 13, 2013
The Canadian economy and Canadian consumers continue to receive intense scrutiny from lenders and policymakers alike. The resilience of the economy and continued rise of house prices in the face of the global recession has puzzled analysts for several years. Forecasts of a U.S.-style collapse in home values and mass charge-offs across consumer credit portfolios have failed to materialize. In this article, we focus on the consumer credit card sector in particular to better understand the trends and consumer behaviours driving outstanding balance growth and payment performance.
Authors: Cristian deRitis, Mustafa Ackay, Stephen Kernytsky
Date: May 09, 2013
In this case study, the client evaluated and quantified the risks and sensitivities of a retail portfolio for which it had limited historical performance data. The analysis showed that even under a very negative macro scenario where the unemployment rate rises up to 16%, the deterioration of the mortgage book would not be dramatic. Analysis of the consumer loan portfolio showed much more sensitivity to the labor market. The impact of other variables, including interest rates and exchange rates, were also evaluated but shown to have a significantly smaller effect on portfolio performance.
Authors: Juan Licari & Jose Suarez-Lledo
Date: March 15, 2013
Econometricians need to build models for forecasting or prediction, strucutural analysis or hypothesis testing, and policy or shock analysis. Each of these applications has an underlying loss or risk function that governs how the model should be built and the properties the preferred model specification should retain. In this paper, we condiser stress-testing, which we view as somewhat distinct from other types of econometric analysis.Author: Tony Hughes Date: June 4, 2012