One of the first effects that the COVID-19 pandemic has had on outstanding U.S. auto loan asset-backed securities (ABS) has been the spike in extensions. In the prime segment for March, extensions on a dollar basis equaled 3.94% of outstanding loans as of the beginning of the month, 12 times February's level of 0.33%. For the four subprime auto loan ABS shelves (Santander's DRIVE and SDART, AmeriCredit, and World Omni's Select), extensions more than quadrupled to 6.82%, from 1.53% in February.

COVID-19 Slows Traffic, But Accelerates Extensions

The spike in extensions isn't surprising given that over 10 million Americans applied for unemployment benefits in March due to layoffs and furloughs as a result of the stay-at-home orders designed to slow the spread of the coronavirus. Most auto lenders have responded by providing affected borrowers payment extensions/deferrals of between 30 and 120 days. These extension/deferral plans differ by lender, with some requiring that the obligor be current at the time of the extension and others allowing the obligor to be up to 90 days' delinquent. Lenders generally waive extension fees, although interest continues to accrue through the extension period. These extensions are intended to be short-term (i.e., to tide the obligor over until their paychecks resume). When used effectively, they can generate lower losses than repossessing and liquidating the vehicles.

Chart 1

Chart 2

Extensions In Prime Pools Are As High As Subprime Pools In Some Cases

While one would expect the level of deferrals to be higher in subprime shelves than in higher-quality pools--that is generally true--it isn't always the case. For example, extensions in California Republic Bank's nonprime pools and those of Ally exceeded those on Santander's SDART platform, and extensions from Ford Credit, Nissan, World Omni, and Mercedes equaled or exceeded those from AmeriCredit.

Chart 3

Chart 4

It's hard to conclude too much though from one month of performance. Further, the banks and captive finance companies may be more highly motivated to offer extensions to their customer base due to the seemingly stronger credit quality of their obligors as well as the lenders' desire to maintain customer loyalty. In addition, given the higher annual percentage rates (APRs) on subprime auto loans, it may be more prudent for those customers to make the payments when due if possible rather than extend the loan and have large amounts owing at the end of the loan term due to accrued interest (for more detail, see "Is There Extension Tension in U.S. Subprime Auto Loan ABS?" published Nov. 29, 2018). In contrast, many prime customers benefit from subsidized APR financing of 0.0%-1.9%, so there is little-to-no additional interest that will accrue during the extension period.

DRIVE Deals Account For Approximately 50% Of Subprime Extensions

GM Financial's AmeriCredit shelf accounts for approximately 29% of the total loans attributable to the four subprime shelves that provide Reg AB II data. Santander's SDART and DRIVE platforms comprise most of the remainder at 28% and 41%, respectively. AmeriCredit's share of extensions by loan count, however, was only 18%, while, perhaps not surprisingly, the deep subprime DRIVE platform accounted for 50%.

Chart 5

Chart 6

Ally Accounts For A Disproportionate Share Of All Prime Extensions

CarMax (17%), Honda (16%), and Toyota (16%) together account for approximately 49% of the total available March loan count for the prime shelves reporting under Reg AB II. However, they account for only 32% of March's extensions. Ally, in contrast, accounts for 11% of March's auto loans, but 34% of the extensions. In mid-March, Ally announced they would defer payments for auto customers for up to 120 days, which was more liberal than most of their peers. In addition, World Omni's share of total available March loans is approximately 7%, while its share of loans in extension is 10%. World Omni's pools tend to have a 47%-50% concentration in Florida, whose travel and leisure economy has been severely affected by shelter-in-place restrictions (see map below). Also, consistent with its AmeriCredit shelf, GM Financial's prime GMCAR shelf's share of extension loans (1%) is smaller than its share of total available March loans (7%).

Chart 7

Chart 8

Nevada And Florida Have Highest Extension Levels

As the two heat maps below indicate, the states with the highest extension levels (based on Reg AB II data) in both the prime and subprime segments are Nevada and Florida. In prime, approximately 8% and 7% of the loans in Nevada and Florida, respectively, were extended in March. In subprime, approximately 11% and 10% of the loans in these two states, respectively, were extended. In Nevada, the leisure and hospitality industry (including casinos) directly employ one-in-four workers in the state, and the casinos have been closed since mid-March.

Chart 9

Chart 10

In our view, extensions will likely remain high for at least the next two-to-three months due to high unemployment levels, and lenders preferring to work with their customers rather than trying to repossess and sell vehicles into a depressed and only partially functioning used vehicle wholesale market. Further, many states have imposed restrictions on collection efforts to a greater or lesser extent, and banned repossessions and the garnishment of CARES Act payments while their state is under a state-of-emergency.

We believe these extensions, by themselves, will not lead to timely interest payment defaults on S&P Global Ratings-rated auto loan ABS transactions due to significant liquidity in those transactions, including reserve accounts, the ability to use principal collections to pay bond interest, and conservative assumptions used when setting legal final maturities (see "The Potential Effects Of COVID-19 On U.S. Auto Loan ABS," published March 26, 2020). However, higher credit losses, including losses on accounts that are extended and subsequently default, could lead to rating downgrades and defaults depending upon the level of credit deterioration and credit enhancement supporting the various classes. Currently, in our view, the transactions that are at greatest risk of downgrade and default are subprime auto loan ABS with noninvestment-grade classes. In the coming months, we will be monitoring extensions and other performance variables on outstanding auto loan ABS transactions and will be taking rating actions as deemed appropriate.

S&P Global Ratings acknowledges a high degree of uncertainty about the rate of spread and peak of the coronavirus outbreak. Some government authorities estimate the pandemic will peak around midyear, and we are using this assumption in assessing the economic and credit implications. In our view, the measures adopted to contain COVID-19 have pushed the global economy into recession (see our macroeconomic and credit updates here: www.spglobal.com/ratings). As the situation evolves, we will update our assumptions and estimates accordingly.

Appendix: Data Notes

The extension rate is calculated as the dollar amount of the loans extended divided by the total beginning balance of the loans for that month.

The data provided for most of the statistics in this article were generated from the Reg AB II loan level filings whether or not the underlying deals were rated by S&P Global Ratings. In the prime segment, we included the securitizations of 14 different originators: Ally Bank, American Honda, California Republic Bank, CarMax, Capital One, Ford Credit, GM Financial, Harley Davidson Credit Corp, Mercedes-Benz Financial Services, Nissan Motor Acceptance, Toyota Motor Credit, USAA Federal Savings Bank, VW Credit, and World Omni Financial (while we classify California Republic's two pools in this analysis, 2017-1 and 2018-1, as nonprime, for ease, we've grouped them in the prime segment). In addition to the subprime shelves of AmeriCredit, Santander, and World Omni Select, whose performance data are filed with the SEC, we also compiled extension data for most of the 144-A subprime securitizers using their monthly servicing report data, as shown in the Subprime - Extension Rate chart above.

The authors would like to thank Bushra Dawawala, Aakansha Khandelwal, Rakesh Mourya, and Sahayajayaseelan Senathikagu for their contributions.