• A MUST WATCH Co-Sponsored Webinar with CU Sol:

    Strategic Planning in a Pandemic
    August 27, 2020

  • COVID-19 Crisis:

    Stress Test and Peer Comparision Report

  • A MUST WATCH Webinar:

    When the Government Lifeline Ends
    July 16, 2020

  • Prepare for COVID-19

    Learn More:
    Stress Testing for Pandemics

Increasing Risk of a W-Recession

Increasing Risk of a W-Recession

Written by: Joseph L. Breeden, PhD | Posted on: | Category:

Creating economic scenarios through the COVID-19 crisis continues to be a challenge. In the last week we had new economic data reporting on conditions three weeks ago heralding the creation of 4.8 million new jobs. At the same time, reports of skyrocketing coronavirus cases are causing businesses to re-close and some of those same newly rehired people to be laid off again. We will not know for another month how many people are losing their jobs again, but continued job growth seems doubtful for a while.

Of course, we warned about this possibility months ago. Slow and steady wins this race to simultaneously control coronavirus and restart the economy. Our W-scenario was specifically designed with the thought that if we reopen too soon and too fast, coronavirus cases could rebound and the economy could be forced closed again. Trying to force a V-shaped recovery without a solid plan for combatting the virus is the method for creating a W-recession instead.

We’re not the only ones who think so. The Federal Reserve released its “Assessment of Bank Capital during the Recent Coronavirus Event” on June 12, 2020. Their V, U, and W scenarios are almost identical to what we have been using. Before their release, we thought that we would add a set of supplementary scenarios to fill out the needed DFAST variables from the Fed’s V, U, and W scenarios. Clearly, that won’t be necessary. As we look forward to future revisions, how states handle the coronavirus resurgence will be key.

Data Quality

As reported last month, employment data quality continues to be an issue. The Bureau of Labor Statistics again reported that some people are being incorrectly classified in their reports. The reported 11.1% unemployment rate could have been 12.3% if potentially misclassified people were included. The BLS does not alter the data as reported from the field. For purposes of creating the best scenarios possible, we do. Rather than the official numbers, we have been using the adjusted unemployment rate since March.

Where some key statistics have not been reported for several months, we attempt to estimate those unreported values using measures similar to the official DFAST variables. Ironically, one of the variables most out of date is the US Index of Leading Economic Indicators. It’s not very “leading” if it hasn’t been reported since February.

The Next Big Uncertainty

Real (and nominal) disposable personal income (DPI) experienced a dramatic 13.6% increase in April via the combination of assistance checks to consumers and increases in unemployment benefits. It dropped in May by 5% with no more assistance checks, but still represents a dramatic increase in consumer financial status.

Historically, sudden increases like that in April are followed by a roughly equal drop the following month. Our models initially predicted that snap-back in May, but when the drop was only partial, the models took that as the new baseline level for the latest scenarios.

We fully expect that politicians will not continue these benefits indefinitely, but we don’t know when it will end. If some businesses are forced to re-close, congress may finally move to extend these assistance programs. In an election year, it seems logical that such an extension would go at least through November. Our models don’t know that it’s an election year or how the next congress will respond to the crisis. Neither do we. Therefore, we have not done any kind of override on the model’s prediction that Real DPI continues from this new baseline level under all three recession scenarios. When the future of government assistance becomes clear, we may incorporate that into the scenarios.

Scenario Details

The most recent economic data does not require significant changes in our scenarios. The V and W scenarios already aligned with these updates. The U is revised to carry forward from current unemployment levels.

“V” Scenario

A V-shaped recession assumes that by mid-summer we see a dramatic wave of businesses reopening and employees being rehired. This would be driven by a resurgence of consumer confidence. That can happen if the following proves to be true: (1) the asymptomatic population proves to be so large that we are already more immune than we realized. A recent Swedish study reinforces this view. (2) Dexamethasone continues to reduce mortality and ICU visits by one third. (3) Hospitals are prepared now for the level of cases that come in. (4) Many businesses are able to reopen with a low-density business model or other innovations. All of these are true to some extent, so a V is possible, although it's difficult to say how much so.

Recent trends in Florida, Texas, Arizona, and many others bring the “V” into doubt. Some state governments and the general public tried to force a V, but a U might now be the best case.

“U” Scenario

A U-shapred recession happens if the previous points are all partly true. The summer is mostly flat as businesses and medical professionals work through what is needed and consumers regain some confidence. Q3 sees a gradual beginning and Q4 is when real growth begins. For the U scenario, we incorporated the April improvement but assume that businesses reopenings will stall for through the end of the year. The result is a long, slow recovery. This scenario is likely given current trends in new coronavirus cases.

“W” Scenario

A W, L, or checkmark recovery is where we struggle to put these pieces in place. If the reclosing now hitting some states becomes severe, more businesses will have to close permanently.

The greatest harm from the reclosings could be to consumer confidence, making it even more difficult for businesses to re-reopen in the future. Many local businesses will find that a low-density business model does not pay the rent. Those business failures could lead to a further drop in commercial real estate prices and defaults on CRE loans. (See Green Street Commercial Property Index as a more real-time example of what is happening there.) This drop in CRE Index is captured to varying degrees in all of our scenarios. The system will eventually self-heal with new business models at lower costs per square foot, but that process takes us well into 2021.


The revised-V, U, and W scenarios are all in play at this point. As always, we know that COVID-19 will spread. We know that our hospitals are better prepared than before. What we cannot predict is the delicate balance between consumer enthusiasm for normalcy and spread of the disease. Too much of either is bad for the economy. We will continue to monitor this balancing act closely.

Software Delivery On Monday, July 6th, we will make these three scenarios available to our clients: COVID-19 V 20200706, COVID-19 U 20200706 and COVID-19 W 20200706 Software users will see the scenarios appear in the software. Please contact us if you have any questions.


Learn More About How DFA's Software Can Help Your Institution?

Recent Publications

All publications below were authored by DFA's own, Dr. Joseph Breeden

  • Image

    Sample Chart

    Reserves: All Loans vs. RE Loans


    CECL (Current Expected Credit Loss) is the new accounting standard for estimating loss reserves on loan portfolios. The CECL guidance provides a great amount of flexibility in which models are used and a range of other choices that may impact the calculations. This book provides details of a study on how to apply CECL to US mortgage data. It seeks to disclose as many modeling details, results, and validation tests as possible so as to provide a reference for comparison and best practices. Because CECL is so similar to IFRS 9 Stage 2, this can also serve as a benchmark for implementing the new international account standards. The book is organized into three parts. Part I: Study Summary provides an overview of CECL, the design of the mortgage study, and the key comparative results across the models tested. Part II: Model Details provides in-depth discussions of how the models were designed and estimated, the coefficients, and the validation. Part III: Background provides additional conceptual material. Chapters 11 and 12 may be particularly useful to those new to modeling, and Chapter 13 puts CECL modeling in the context of lending analytics overall.

    • Date // May 2018

    Sample Chart

    Vintage Performance


    Building on the solid foundation of the previous bestselling first impression, this extended updated impression walks through the various issues of retail lending and develops approaches to address the interaction between economic cycles and retail lending. The complexity of time is extensively explored: vintages, current time and maturity. Reinventing Retail Lending Analytics, Second Impression covers complex issues such as scenario based forecasting, stress testing, volatility analysis, economic capital and portfolio optimisation, credit scoring and last, but not least, model risk.

    The book ends by providing examples of the application of nonlinear decomposition. These examples will provide you with rich data sets for exploring portfolio dynamics and improving portfolio management using nonlinear decomposition techniques.

    • Date // March 2019

    Sample Page



    The new loan loss accounting rules for CECL and IFRS 9 require thousands of organizations to learn about modeling. Likewise, accountants and others in finance are now required to learn about statistical modeling concepts. This book is intended to define terms in a manner consistent with decades of academic literature on statistical modeling and hopefully reduce some of the noise and confusion just around definition of terms. It may also serve as a useful guide to analysts new to the field tasked with IFRS 9 compliance, the international loss accounting rules, and credit risk modeling in general.

    Each chapter of this book is a term that one might encounter when discussing creating lifetime loss forecasting models for CECL or IFRS 9. Not every term is a model, and some models listed are being mentioned only to explain why they are not likely to be used for loss forecasting. The CECL guidelines and subsequent FAQs have given examples of modeling techniques. Some people new to loss forecasting have assumed that those are all the available or applicable methods. This book is meant in
    part to dispel that misconception.

    The definitions and descriptions provided here are meant to provide an intuitive understanding across a range of modeling techniques. Mathematical derivations are kept to a minimum. The references listed will provide all the necessary details for an eager analyst.

    • Date // June 2018
© 2020 Deep Furture Analytics Contact Us