CECL requirements for Macroeconomic Scenarios

CECL requirements for Macroeconomic Scenarios

Hakan Danis, Director of Stress Testing at Union Bank shares his insight on CECL requirements for macroeconomic scenarios.
About Hakan

I am an economist by training and experienced several serious recessions (1994, 1999, 2001 in Turkey and 2008 in the U.S.) during turning points of my life and drew my career path. I am holding a PhD in economics, focusing on applied econometrics and monetary economics which landed me the Senior Economist position in BBVA. At BBVA, I was responsible for all U.S. macroeconomic projections and regularly writing on topics such as the U.S. monetary policy, inflation etc. And, even before SCAP or DFAST, I was designing baseline, stress and alternative macroeconomic scenarios for the BBVA Group. When the Bank became subject to CCAR, it was easy for us to fulfill regulatory requirements related to Scenario Design. In 2012, I started working for MUFG Union Bank where I am mainly responsible for the macroeconomic modeling and scenario design. In particular, I design macroeconomic scenarios for the Intermediate Holding Company which requires projecting 130+ series for each scenario for use in CCAR, Mid-cycle DFAST, ICAAP, RRP, JFSA and monthly budget baseline scenarios. We also do the expansion of the supervisory scenarios in house and I will be responsible for CECL going forward. I am also a member of Overlay Committee which approves any overlays in any of the model results (PPNR, credit etc.) during DFAST and CCAR.

The answers provided are my personal opinion and do not reflect the opinion of MUFG Union Bank or its subsidiaries.

When discussing CECL requirements for Macroeconomic Scenarios, can you please explain how CCAR processes could be leveraged? And what are the pros to this being done successfully.

In my experience, first of all, most of the CCAR Banks would prefer to use CCAR processes whenever it is applicable because these financial institutions have devoted so much people and time and spent so much money. Second, most of their processes are now mature and accepted by their top management and regulators. Therefore, there are not so many unknowns about these processes and it is easier to implement these models and processes for CECL. Third, some of these institutions have even started using these processes as BAU. Hence, in my opinion, CCAR/DFAST institutions will and should leverage CCAR processes.

In your opinion, what should the frequency and timing of scenario design be?

It depends on the institution but it must be at least quarterly. Some institutions might prefer monthly but it would require so much resources and very efficient processes. Institutions must do cost-benefit analysis to decide on the frequency of the macroeconomic scenario.

How can financial institutions best manage governance processes for economic scenario? And have you got any advice for your fellow peers?

Some of the financial institutions might have already been producing baseline macroeconomic projections each quarter. However, in most of the cases, this would still not be enough for CECL purposes. Under CECL, macroeconomic scenario design process needs Sarbanes-Oxley (SOX) controls and therefore, in my personal view, the institutions need to improve their governance processes around scenario design projections. I am expecting that most of the DFAST and CCAR banks will leverage their knowledge and processes from CCAR. However, in certain cases CCAR processes might be too strict and costly. Therefore, each Bank should decide on how to proceed.

At the CECL Congress you will be speaking on your insight regarding – CECL requirements for Macroeconomic scenarios. Why do you believe this is a key talking point in the industry right now and what can risk professionals gain from this insight?

In my personal view, the requirements are new and unclear and the deadline is approaching really fast. There are so many gray areas in the regulations that no one is 100% sure what or how they should do certain things to comply with the CECL requirements. This conference will cover many of these areas and we will have an opportunity to discuss with our peers and learn different insights. For example, from macroeconomic scenario perspective, there are crucial questions that the institutions must answer before starting their CECL processes:
(1) What is “reasonable and supportable” forecasting period? Is it 1 year, 9Qs or 5 years? How will institutions decide?
(2) What about periods after the reasonable and supportable period? How long should be the forecasting window, in total?
(3) Would leveraging CCAR processes be enough or too much? How should the governance process be for macroeconomic scenario?
(4) What will be the frequency of scenario design?
(5) How many scenarios should be used? Do institutions need macroeconomic models for these scenarios?
(6) How should institutions communicate to the public, especially if there are changes in their macroeconomic scenarios?
I will cover these points in the conference and participants, including me, would benefit from discussions with other participants.

What does the future look like for the Stress Testing Risk Professional right now? And what changes or problems do you see in the horizon and theoretically, how would you tackle these?

In the U.S., with the new presidency, CCAR/DFAST rules are under question and that’s why the future of Dodd-Frank Act and stress testing is uncertain. Even if some of the proposals pass the Congress and are enacted, there will be less stress test requirements and therefore, demand for Stress Testing Risk Professionals like us will decrease for sure. However, most of the financial institutions accept the benefits of stress testing and have started to use stress testing in their BAU decisions. This will keep interest for Stress Testing Risk Professionals. In addition, there are other regulations keep the demand high. In the next couple of years, financial institutions will be trying to fulfill CECL rules where most of the opportunity will be.

Haken will be presenting at the CECL Congress 2017 on CECL requirements for Macroeconomic Scenarios, including insight on: What is ‘reasonable and supportable’ forecasting period, can CCAR be leveraged, how many scenarios should be used and more.
//]]>