articles Ratings /ratings/en/research/articles/231114-u-s-auto-loan-abs-tracker-september-2023-performance-12911882 content esgSubNav
In This List
COMMENTS

U.S. Auto Loan ABS Tracker: September 2023 Performance

COMMENTS

SF Credit Brief: U.S. CMBS Delinquency Rate Rose 10 Bps To 5.3% In October 2024; Office Rate Surged Past 9.0%

COMMENTS

Analytical Approach: EU Taxonomy Assessment

COMMENTS

Japan Private-Sector RMBS Performance Watch: Stable Labor Market Provides Support

COMMENTS

JHF RMBS Performance Watch October 2024: Delinquency Rise Will Be Negligible


U.S. Auto Loan ABS Tracker: September 2023 Performance

U.S. auto loan asset-backed securities (ABS) composite performance weakened in September 2023, with annualized losses increasing in the prime and subprime segments both month over month and year over year. Prime losses returned to September 2019 levels, while subprime losses rose to the highest September level since 2009. Modified subprime composite losses (which exclude the three large deep subprime issuers) reached a record high September level. Recoveries for prime and subprime composites both continued to weaken, declining month over month and year over year, while, 60-plus-day delinquencies stabilized for prime and declined for subprime.

On a static pool basis, S&P Global Ratings is seeing a divergence in performance between the prime and subprime segments. Prime issuers are generally reporting stable to lower cumulative net losses (CNLs) relative to pre-pandemic vintages, with just a few exceptions, while most subprime issuers are reporting higher-than-historical losses. The greater deterioration in subprime losses reflects the disproportionate impact inflation has had on the lower income borrower segments. However, the first-quarter 2023 subprime vintage is showing some early improvement relative to the 2022 vintage.

The deterioration partly reflects the increase pressure on subprime consumers, who are having a more difficult time juggling inflationary pressures than their higher income prime counterparts, especially with monthly car payments increasing by up to 25% in the past two years. Seventy-five percent of the subprime issuers we rate that securitized in both 2016 (the last annual vintage that didn't benefit from COVID-19-related aid) and 2022 are reporting higher CNLs on the 2022 vintage.

Prime And Subprime Losses Rose Month Over Month; Subprime Reached Highest September Level Since 2009

Prime annualized net losses increased to 0.57% in September 2023 from 0.54% in August 2023 and 0.38% in September 2022. This loss deterioration exhibits the cumulative effect of lower recoveries, diminishing pandemic-related savings, and inflationary pressures. Nevertheless, prime annualized net losses are still at par with the September 2019 pre-pandemic levels.

Subprime annualized net losses reached the highest September levels since 2009, increasing to 9.14% in September 2023 from 8.98% in August 2023 and 7.55% in September 2022. However, unlike prime, subprime annualized net losses exceeded the September 2019 pre-pandemic levels of 8.57%. The subprime issuers with the three highest increases in annualized net losses month over month were Prestige, which reported a 264 basis points (bps) increase to 12.14% from 9.50%, Tidewater (264 bps; to 5.10% from 2.46%), and United Auto (224 bps; to 23.76% from 21.52%).

Modified subprime annualized losses (after netting out these three large deep subprime issuers) reached the highest September levels since we created the modified index in 2014, with losses rising to 7.98% in September 2023 from 7.86% in August 2023 and 5.97% in September 2022.

The deterioration in the subprime and modified subprime auto loan composites highlights the negative effects inflation is having on lower-income consumers, the ease in the underwriting standards by some lenders, and lower recovery levels. Some lenders also attribute the weakness to their credit lending models not sufficiently accounting for the temporary uplift in consumers credit bureau profiles due to COVID-19-related stimulus (also referred to as "FICO inflation").

Table 1

Net loss rate composite(i)
Sep-09 Sep-14 Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 Sep-20 Sep-21 Sep-22 Aug-23 Sep-23
Prime (%) 1.50 0.51 0.50 0.63 0.68 0.61 0.57 0.21 0.17 0.38 0.54 0.57
Subprime (%) (ii) 10.32 7.72 8.07 8.86 8.49 8.56 8.57 3.54 4.11 7.55 8.98 9.14
Subprime modified (%) N/A 7.04 7.08 6.88 6.81 6.55 6.32 3.06 3.03 5.97 7.86 7.98
(i)Represents monthly annualized losses. (ii)Excludes three large deep subprime issuers: American Credit Acceptance, Exeter, and DRIVE. N/A--Not applicable.

Chart 1

image

Recoveries Declined Month Over Month And Below 2019 Pre-Pandemic Levels

Prime and subprime recoveries declined both month over month and year over year. The prime recovery rate slumped to 53.43% in September 2023 from 61.94% in August 2023 and 60.74% in September 2022. CarMax Auto Owner Trust, GM Financial Consumer Automobile Receivables Trust, and Honda Auto Receivables Owner Trust, which represented almost 30% of the prime composite in September, contributed the most to the declines.

We believe recovery rates will continue to decline as transactions that were issued in the second half of 2021 through 2022 become a larger share of outstanding collateral. Vehicles financed in these pools were purchased when prices reached their peak (second half of 2021 to early 2022), and used vehicle values have generally declined since then. The 2019 and 2020 transactions included collateral purchased in the years before the rapid run up in prices (due to vehicle shortages) and benefited from very high recovery rates--often higher than 100%. However, these transactions are paying off and represent a declining share of the outstanding collateral.

Lenders also have been attributing the lower recovery rates to an increase in full-balance charge-offs. Because it's taking longer to "hook" vehicles that have been put out for repossession (repo) orders, those accounts are reaching the delinquency charge-off definition (often 120 days) before the vehicles are sold at auction. This delay is due to the shortage of repo agents since COVID-19 pandemic, and the growing backlog of repos has only exacerbated the situation.

Subprime recovery rate decreased slightly to 38.94% in September 2023 from 39.58% in August 2023 and 44.57% in September 2022. The recovery rate for the modified subprime composite (which excludes the three deep subprime issuers) was nearly unchanged month over month at 39.16% in September 2023, though they remained lower than the 44.14% in September 2022 and 41.48% in September 2019.

Table 2

Recovery rate composite(i)
Sep-09 Sep-14 Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 Sep-20 Sep-21 Sep-22 Aug-23 Sep-23
Prime (%) 53.32 56.31 59.38 57.70 52.85 52.01 53.90 79.91 88.69 60.74 61.94 53.43
Subprime (%) 41.53 40.37 39.95 39.69 39.25 39.42 41.24 57.92 52.03 44.57 39.58 38.94
Subprime modified (%)(ii) N/A 40.83 41.11 40.86 40.88 40.42 41.48 57.25 52.12 44.14 39.09 39.16
(i)Represents monthly annualized losses. (ii)Excludes three large deep subprime issuers: American Credit Acceptance, Exeter, and DRIVE. N/A--Not applicable.

Chart 2

image

Prime And Subprime Delinquencies Rose, Despite Stabilizing Month Over Month

The prime 60-plus-day delinquency rate remained unchanged month over month at 0.54% in September 2023, though it increased year over year from 0.44% in September 2022. The subprime 60-plus-day delinquency rate decreased slightly to 5.99% in September 2023 from 6.04% in August 2023, though it also increased year over year from 5.30% in September 2022. In addition, prime and subprime delinquencies were higher than their September 2019 pre-pandemic levels of 0.49% and 5.27%, respectively. The September 2023 drop in subprime delinquencies was the first monthly decline since April 2023.

Table 3

60-plus-day delinquency rate composite(i)
Sep-09 Sep-14 Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 Sep-20 Sep-21 Sep-22 Aug-23 Sep-23
Prime (%) 0.71 0.39 0.45 0.46 0.45 0.42 0.42 0.33 0.31 0.44 0.54 0.54
Subprime (%) 5.32 4.31 4.75 5.04 4.95 5.36 5.27 3.68 3.71 5.30 6.04 5.99
Subprime modified (%)(ii) N/A 3.80 4.03 3.83 3.62 3.78 3.78 2.69 2.43 3.78 4.96 4.97
(i)Represents 60-plus delinquencies. (ii)Excludes three large deep subprime issuers: American Credit Acceptance, Exeter, and DRIVE. N/A--Not applicable.

Chart 3

image

Extension Rates Declined Month Over Month, But Exceeded The September 2019 And 2022 Levels

Prime and subprime issuers both saw extension rates decline month over month. Prime extensions fell to a weighted average of 0.50% in September 2023 from 0.55% in August 2023, while extensions for subprime public and 144a transactions combined decreased to a weighted average of 3.74% in September 2023 from 3.80%. Nevertheless, extensions for prime and subprime public and 144a transactions combined rose year over year from 0.36% and 3.04% in September 2022, respectively, (see chart 4).

Chart 4

image

Prime extension rates

Of the 15 prime issuers listed in table 4A, only two (Harley-Davidson and Capital One) reported a month-over-month uptick in extensions. The remaining 13 reported declines or stable extension levels. However, there were significant year-over-year increases and growth since September 2019. Specifically:

  • Eleven prime issuers reported year-over-year increases in extensions.
  • Of the 14 prime issuers with September 2019 extension levels, 10 reported higher levels as of September 2023.

CarMax reported the highest increase relative to September 2019 (60 bps; to 1.04% from 0.44%), and Mercedes-Benz reported the second highest increase (54 bps; to 0.72% from 0.18%).

Table 4A

Prime issuer shelf extension (% of balance)
Sep-19 Apr-20 Sep-22 Aug-23 Sep-23
Ally Auto Receivables Trust 0.41 7.49 0.48 0.65 0.65
BMW Vehicle Owner Trust 0.13 7.58 0.12 0.23 0.20
Capital One Prime Auto Receivables Trust 0.02 1.94 0.09 0.06 0.07
CarMax Auto Owner Trust 0.44 7.33 0.72 1.18 1.04
Carvana Prime N/A N/A 0.48 0.47 0.39
Ford Credit Auto Owner Trust 0.80 6.35 0.96 1.04 1.01
GM Financial Consumer Automobile Receivables Trust 0.32 2.27 0.37 0.44 0.43
Harley-Davidson Motorcycle Trust 0.30 4.00 0.34 0.50 0.55
Honda Auto Receivables Owner Trust 0.16 4.51 0.11 0.17 0.16
Hyundai Auto Receivables Trust 0.34 4.22 0.36 0.46 0.36
Mercedes-Benz Auto Receivables Trust 0.18 7.21 0.43 0.80 0.72
Nissan Auto Receivables Owner Trust 0.57 5.57 0.37 0.29 0.27
Toyota Auto Receivables Owner Trust 0.34 6.24 0.24 0.35 0.32
Volkswagen Auto Loan Enhanced Trust 0.16 5.81 0.13 0.23 0.17
World Omni Auto Receivables Trust 0.48 8.82 0.51 0.62 0.58
Prime average(i) 0.33 5.76 0.36 0.55 0.50
(i)The weighted average extension data of the public Reg AB II securitizers noted above. N/A--Not applicable.
Subprime extension rates

Of the 17 rated subprime issuers with both August and September 2023 extension data, only seven reported a month-over-month increase in extension rates. However, there was significant growth in extension levels year over year and since September 2019, indicating that the obligors are having greater difficulty paying all of their financial obligations compared to the prior years. Specifically:

  • Of the 16 subprime issuers with both September 2019 and September 2023 data, 12 reported increased extension levels. The largest increases were observed for Prestige (180 bps; to 4.07% from 2.27%), Exeter (177 bps; to 5.72% from 3.95%), and Flagship (172 bps; to 4.69% from 2.97%).
  • America's Car Mart had the highest deferral and extension rate in September 2023 at 8.52%, and Westlake had the second highest at 6.72%.

This month's data include 60-plus-day delinquencies for the subprime segment because some companies that have high extensions also have low 60-plus-day delinquencies (e.g., Westlake). When we add these two metrics, Tidewater and Exeter had the highest combined level of 60-plus-day delinquencies and extensions at 21.62% and 14.97%, respectively, in September 2023.

Table 4B

Subprime issuer shelf extension (% of balance)
Extensions 60+ day DQs 60+ day DQs and extensions
Sep-19 Apr-20 Sep-22 Aug-23 Sep-23 Sep-23 Sep-23
American Credit Acceptance 3.14 5.15 4.53 4.11 4.22 9.08 13.30
AmeriCredit Automobile Receivables Trust 2.73 6.93 3.00 3.54 3.22 2.15 5.37
America's Car Mart N/A N/A N/A 8.55 8.52 1.22 9.74
Avid Automobile Receivables Trust 1.88 2.82 3.66 2.61 2.96 6.75 9.71
Carvana (subprime) N/A N/A 3.64 1.93 1.97 8.75 10.72
CPS Auto Receivables Trust 3.45 10.21 4.52 5.86 4.83 7.93 12.76
Drive Auto Receivables 2.05 20.64 2.00 1.92 1.88 5.37 7.25
DriveTime Automotive Group Inc.(i) 2.32 8.88 4.21 3.47 3.03 15.20 18.23
Exeter Automobile Receivables Trust 3.95 11.80 5.04 5.84 5.72 9.25 14.97
First Investors Auto Owner Trust 3.57 3.94 2.85 3.11 2.76 5.88 8.64
Flagship Credit Auto Trust 2.97 18.81 3.71 4.92 4.69 8.34 13.03
Foursight Capital Automobile Receivables Trust N/A N/A N/A 2.74 2.80 2.31 5.11
GLS Auto Receivables Issuer Trust 4.16 11.39 3.21 3.61 3.54 8.11 11.65
Prestige Auto Receivables Trust 2.27 6.24 4.54 4.06 4.07 6.10 10.17
Santander Drive Auto Receivables Trust 1.56 17.28 1.65 1.68 1.70 5.37 7.08
Tidewater Auto Receivables Trust 1.66 8.36 1.36 1.61 1.23 20.39 21.62
United Auto Credit Securitization Trust 4.31 6.02 4.27 5.86 5.04 4.22 9.26
Westlake Automobiles Receivables Trust 5.42 7.41 6.00 6.33 6.72 1.38 8.10
World Omni Select (subprime)(ii) 0.74 19.58 1.80 1.19 1.15 2.66 3.81
World Omni Select (nonprime)(iii) N/A N/A N/A 1.99 1.78 2.31 4.09
Subprime average(iv) 2.88 12.75 3.04 3.80 3.74 6.92 10.21
(i)The May and June 2023 extension rates for DriveTime Automotive Group Inc. include only the series 2021-3 through 2023-2 transactions. These are the only transactions with at least four months of performance in which the servicing reports provide extension info on a dollar basis (as opposed to account-based). The company previously provided dollar-based extensions through Dec. 31, 2021, for all of its outstanding transactions from August 2019 through December 2021. (ii)World Omni Select (subprime) inlcudes WOSAT 2019-A and 2020-A. (iii)World Omni Select (nonprime) inlcudes WOSAT 2021-A and 2023-A. (iv)For extensions, the weighted average of all the issues in the respective month. For 60+ day DQs and 60+ day DQs and extensions, the simple average. DQs--Delinquencies. Carvana--Carvana Auto Receivables Trust. WOSAT--World Omni Select Auto Trust. N/A--Not applicable.

Auto Loan Static Indexes (ALSI)

Prime: CNLs still tracking lower than pre-pandemic levels

Prime CNLs for the 2022 vintage are 0.23% at month 10 (see chart 5). This is lower than the 0.27%-0.30% levels for the 2016-2019 vintages at the same month, but higher than the 0.18% and 0.17% levels for the 2020 and 2021 vintages, respectively. The first-quarter 2023 vintage is tracking similarly to 2022, and prime cumulative gross losses (CGL) are showing trends similar to CNLs through first-quarter 2023 vintage. (See chart 6 for CGL data and charts 15-45 for issuer-specific CNL performance.)

Cumulative recoveries on the 2022 vintage, which got off to a weak start, finally crossed both the 45% threshold and 2008's depressed recovery level of 44% at month 10 (see chart 7). Over time, we would expect liquidations on previously-charged off loans to lift the average cumulative recovery rate to 50%. The initial weakness is likely due to the drop in used vehicle values (which started to decline in early 2022), especially if the repossessed units were new vehicles purchased from mid-2021 to year-end 2022 that were priced above the manufacturer's suggested retail price (MSRP). This "market-based" pricing is gradually being eliminated, causing vehicles purchased during this time frame to experience greater declines in value. We've also observed a noticeable decline in the cumulative recovery rates for most captives between their first-quarter and fourth-quarter 2022 securitizations. However, the first-quarter 2023 securitizations are reporting much higher cumulative recovery rates of 45.6% at month 7, compared with 38.4% for the 2022 annual vintage.

The 2022 vintage is also showing weakness on the delinquency front, reporting higher 60-plus-day delinquencies, which at month 10 are comparable to the levels observed for the 2008 vintage (see chart 8). However, losses have not risen to the same extent because, despite the hardship that caused obligors to miss some payments, many have been able to resume their payments. But they haven't been able to bring their accounts current, which would require their making two or more payments within a month (a phenomenon frequently referred to as "churning in delinquency status.") This is understandable, given that the average monthly new vehicle payment has increased by more than 25% in the past two years to approximately $730 from $580. First-quarter 2023 delinquencies are trending higher as well.

Chart 5

image

Chart 6

image

Chart 7

image

Chart 8

image

Subprime: 2022 vintage CNLs remain higher than historical levels, but first-quarter 2023 vintage is off to a better start

The 2022 subprime vintage is reporting worse-than-historical loss levels, with CNLs of 4.39% in month 10, compared with 3.82% for 2008 and 3.66% for 2016 (see chart 9). The 2022 subprime vintage, unlike the 2020 and 2021 vintages, was unaffected by COVID-related stimulus. The second- and fourth-quarter 2022 subprime vintages are showing the most deterioration of the quarterly 2022 vintages. The higher CNLs are due to higher defaults, a precipitous drop in recoveries, the inflationary impact on these lower income obligors, and, in our view, the liberalization of credit standards among some lenders and/or their underwriting methods' inability to account for pandemic-related FICO inflation. The first-quarter 2023 subprime vintage is performing noticeably better with CNLs of 2% at month 7, which is well below the 2.6% average in 2022 and in line with 2016 levels.

Although the 2022 subprime vintage is performing worse than 2008 on a CNL basis, it is paying down more quickly than the 2008 vintage. As a result, despite having a similar pool factor of approximately 72%, the 2022 vintage has experienced a lower CNL of 4.39%, compared with the approximately 4.80% for the 2008 vintage (see chart 10). This could imply that issuers are taking more front-loaded losses than during the Great Financial Crisis of 2008. However, a lot depends on future unemployment levels and inflation.

The subprime market has changed since 2008, with more issuers now focusing on deep subprime lending. When looking at our modified subprime index, which excludes the three largest deep subprime lenders, we note that although CNLs for the 2022 vintage are trending lower than the 2008 level, they are still higher than all vintages since 2016 (see chart 11). The performance deterioration in 2022 occurred despite the absence of issuance from Santander's deep subprime shelf (Santander Drive Auto Receivables Trust). Santander's retreat contributed to deep subprime issuance declining to 20% of our subprime index in 2022 from 33% in 2021. This was more than offset by the dramatic increase in losses in Exeter's and ACA's 2022 securitizations. (For the composition of our subprime index, see "U.S. Auto Loan ABS Tracker: June 2023 Performance," published Aug. 15, 2023).

Cumulative recovery rates for the 2022 vintage remain depressed at 31.89% at month 10, compared with 33.91% for the 2008 vintage and 35.43% for the 2016 vintage (the last vintage unaffected by COVID-stimulus; see chart 13). Subprime recoveries have improved slightly for the first-quarter 2023 vintage increasing to 28.07% at month 7, compared with 26.65% for the 2022 vintage. While some issuers attribute the lower recovery rates to more full-balance charge-offs because it's taking longer to find and repossess vehicles, falling used vehicle prices in the secondary market are also contributing to the decline.

Sixty-plus-day delinquencies have ticked up for all vintages, with the 2021 vintage reporting 6.09% at month 22, which is higher than those for the 2007, 2008, and 2015-2020 vintages. However, the first-quarter 2023 vintage is showing an improvement with 60-plus-day delinquencies averaging 3.13% at month 7, which is down from 4.64% for 2022 at the same month and in line with the 2015 and 2016 delinquency levels. Note that we included the 2007 vintage in this analysis because it has higher delinquencies than the 2008 vintage, despite having lower CNLs (see chart 14).

Chart 9

image

Chart 10

image

Chart 11

image

Chart 12

image

Chart 13

image

Chart 14

image

Issuer-Specific CNL Deterioration

The confluence of adverse factors affecting the consumer has, in many cases, resulted in higher CNLs for the 2022 vintage pools versus the 2016 vintage pools. We derive our base-case CNLs on an issuer-specific basis, and thus analyze performance at the issuer level. Below we provide additional performance details for prime and subprime issuers.

Prime issuers are reporting mixed performance results on their 2022 auto ABS

In the prime segment, most issuers are still reporting stable to better performance on their 2022 and first-quarter 2023 securitizations than the 2016 and 2017 vintages, which were largely unaffected by COVID-19 stimulus. These issuers include Ally, BMW, Ford Credit, GM Financial, Honda, Hyundai, Nissan, Toyota, and World Omni. However, certain issuers are reporting higher losses on their 2022 securitizations than their 2016 or 2017 ones. These companies include CarMax, Mercedes, Harley, and USAA (although USAA continues to perform within our original expectations).

Given the deterioration in performance for CarMax and Mercedes, we recently revised and raised our loss expectations for their rated securitizations. In October 2023, we reviewed 13 CarMax transactions and raised our expected CNLs (ECNLs) on the transactions issued in 2022 (see the Related Research list below). In September 2023, we reviewed four Mercedes transactions and raised our ECNLs on the MBART 2021-1, 2022-2, and 2023-1 transactions (see the Related Research list below).

Chart 15

image

Chart 16

image

Chart 17

image

Chart 18

image

Chart 19

image

Chart 20

image

Chart 21

image

Chart 22

image

Chart 23

image

Chart 24

image

Chart 25

image

Chart 26

image

Chart 27

image

Chart 28

image

Chart 29

image

75% of subprime issuers have higher losses on their 2022 auto ABS than their 2016 vintages

In the subprime segment, nine of the 12 issuers that securitized in both 2016 and 2022 are reporting higher CNLs on the 2022 vintage. These issuers are American Credit, CPS, Exeter, First Investors, Flagship, Foursight, Prestige, SDART, and UACST. The three issuers reporting better-than-historical performance are AmeriCredit, Westlake, and DriveTime Automotive Group Inc.

Since the beginning of this year, we have revised and increased our ECNLs for many subprime issuers. In October alone we raised our ECNLs on Prestige 2022-1, CPS 2022-C, and Foursight 2022-2 (see the Related Research list below). Prestige has experienced significant deterioration, largely due to the reduction in loans to previously bankrupt obligors (for Prestige these loans have historically performed better than loans to nonbankrupt obligors). We also placed our 'BB- (sf)' rating on Prestige 2022-1's class E notes on CreditWatch with negative implications (see the Related Research list below). For Foursight and CPS, the weakness continued in the first-quarter 2023 vintages.

Chart 30

image

Chart 31

image

Chart 32

image

Chart 33

image

Chart 34

image

Chart 35

image

Chart 36

image

Chart 37

image

Chart 38

image

Chart 39

image

Chart 40

image

Chart 41

image

Chart 42

image

Chart 43

image

Chart 44

image

Chart 45

image

Chart 46

image

Chart 47

image

Collateral Characteristics: Higher Credit Scores; Credit Unions Crossed the 84-Month Barrier

Prime

For the nine months ended September 2023, prime transactions have generally seen originations with longer terms and higher FICO scores (see table 5A). The significant changes compared with full-year 2022 include:

  • The weighted average FICO increased to 759 from 754--the highest in recent history.
  • The weighted average annual percentage rate (APR) increased to 5.46% from 4.41%--the highest level since 2009.
  • Used vehicles as a percent of originations decreased to 33.73% from 35.13%.
  • Loans with original maturities of 73-84 months increased to 16.19% from 13.17%.
  • The weighted average loan-to-value (LTV) ratio increased to 96.97% from 95.95%.

In addition, two credit unions also securitized loans with original term of 85-87 months: GECU (34.5% of its series 2023-1 transaction and Oregon Community Credit Union (5.95% of its series 2023-1 transaction). We believe longer-term loans are supporting the growth in new vehicle financing, especially since higher interest rates and elevated vehicle prices are driving up monthly payments.

Table 5A

Prime collateral trends(i)
WA APR (%) Used (%) Loans with original terms > 60 months (%) Loans with original terms of 61-72 months (%) Loans with original terms of 73-75 months (%) Loans with original terms of 76-84 months (%)(ii) Loans with original terms of 73-84 months (%)(ii) Loans with original terms of > 84 months (%)(iii) WA original maturity WA LTV ratio (%) WA FICO score
2016 3.51 29.75 50.15 45.00 - - - - 64.32 96.94 746.29
2017 3.59 28.38 55.92 48.16 - - - - 65.52 95.94 748.67
2018 3.73 27.66 58.14 52.35 - - - - 65.80 95.81 754.94
2019(ii) 4.50 29.66 60.51 52.54 7.01 0.95 7.96 - 66.16 95.64 754.33
2020 4.24 29.41 61.81 54.69 6.39 0.73 7.12 - 66.25 96.66 757.00
2021 4.29 36.40 63.06 52.97 8.48 1.61 10.09 - 66.73 96.16 754.00
2022 4.41 35.13 66.87 53.65 10.45 2.77 13.22 - 67.17 95.95 754.00
2023(iv) 5.46 33.73 68.12 50.91 9.64 6.55 16.19 0.24 67.77 96.97 759.00
(i)The collateral characteristics of revolving transactions are not included because the collateral pools are dynamic. The 2016-2020 vintages include most term transaction not rated by S&P Global Ratings. (ii)Effective August 2019, we began reporting loans with terms between 73-75 months and 76-84 months.(iii)In 2023, GECU 2023-1 and OCCU 2023-1 included more than 84 months. (iv)As of Sept. 30. WA--Weighted average. APR--Annual percentage rate. LTV--Loan-to-value. GECU--General Electric Credit Union Auto Receivables Trust. OCCU--OCCU Auto Receivables Trust.
Subprime

The collateral characteristics for subprime transactions issued in the nine-months ended September 2023 exhibited an increase in longer-term loans and higher FICO scores (see table 5B). Notable observations relative to full-year 2022 include:

  • The weighted average APR surged to approximately 18.82% from 16.69% as most lenders have increased their APRs, given higher ABS funding costs.
  • The percentage of loans with original terms of 73-75 months and 76-84 months increased to 13.20% and 3.10%, respectively, from 12.60% and 2.27%.
  • The weighted average FICO score increased to 595 from 593.
  • The percentage of used vehicle financing increased to 85.00% from 84.75%
  • The weighted average LTV increased to 112.20% from 110.30%. This is a negative trend, though used vehicles generally have higher financed LTVs than new vehicles.

We note that the statistics in table 5B do not capture the extent to which lenders have improved their mix of originations because they do not track the issuers' proprietary scores, which are often better predictors of performance than FICO scores only. We also noticed that several lenders' 2023 pools have migrated to a stronger composition of borrowers, and, in some cases, lenders have eliminated or greatly reduced certain high-loss programs.

Table 5B

Subprime collateral trends(i)
WA APR (%) Used (%) Loans with original terms > 60 months (%) Loans with original terms of 61-72 months (%) Loans with original terms of 73-75 months (%)(ii) Loans with original terms of 76-84 months (%)(iii) Loans with original terms of 73-84 months (%)(iii) WA original maturity WA LTV ratio (%) WA FICO score
2016 16.85 68.25 83.27 79.56 - - - 68.52 112.55 575
2017 17.79 69.05 84.61 79.28 - - - 68.94 110.57 578
2018 17.97 66.53 83.03 74.55 - - - 68.65 110.28 587
2019(ii) 17.84 70.55 82.63 74.49 8.14 - - 68.69 112.45 584
2020 17.21 72.42 83.04 73.15 9.48 0.43 9.91 68.98 111.87 588
2021 17.07 78.65 86.3 76.25 8.56 1.46 10.02 69.68 109.8 590
2022 16.69 84.75 85.64 70.13 12.6 2.27 14.87 69.75 110.3 593
2023(iv) 18.82 85.00 84.46 68.13 13.2 3.1 16.34 69.62 112.2 595
(i)The collateral characteristics of revolving transactions are not included because the collateral pools are dynamic. The 2016-2020 vintages include most term transaction not rated by S&P Global Ratings. (ii)The 2023 vintages include 73-78 months for EART (series 2023-1 through 2023-4) and FCRT (series 2023-1 and 2023-2). (iii)Effective August 2019, we began reporting loans with terms between 73-75 months and 76-84 months. (iv)As of Sept. 30. WA--Weighted average. APR--Annual percentage rate. LTV--Loan-to-value. EART--Exeter Automobile Receivables Trust. FCRT--Foursight Capital Automobile Receivables Trust.

Auto Loan ABS Rating Activity/Revised Loss Expectations

In October 2023, we raised 52 ratings, affirmed 85 ratings, and placed one rating on CreditWatch negative. This brought the total number of auto loan ABS-related upgrades, downgrades, and affirmations to 305, six, and 394, respectively, as of Oct. 31 (see tables 6-8). We also reviewed and revised our loss expectations for 32 transactions: we raised seven, lowered 21, and maintained four (see tables 9-16).

Table 6

Surveillance actions
Rating action (no. of classes) Expected cumulative net losses (no. of transactions)
Issuer Date Transaction reviewed Upgrades Downgrades CreditWatch Affirmations Increased Decreased Maintained
Prime
FCAOT 2022-C Oct. 10, 2023 1 1 - - 5 1 -
CAOT Oct. 13, 2023 13 20 - - 49 4 7 2
CRVNA 2022-P3 Oct. 19, 2023 1 2 - - 4 - - 1
USAOT 2022-A Oct. 20, 2023 1 1 - - 3 - 1 -
Subprime
FCRT 2022-2 Oct. 02, 2023 1 3 - - 2 1
CPSART Oct. 13, 2023 9 15 - - 8 1 7 1
PART 2022-1 Oct. 27, 2023 1 2 - 1 3 1 -
AMCAR Oct. 31, 2023 5 8 - - 11 - 5 -
Total 32 52 0 1 85 7 21 4
FCAOT--Ford Credit Auto Owner Trust. CAOT--CarMax Auto Owner Trust Transaction. CRVNA--Carvana Auto Receivables Trust. USAOT--USAA Auto Owner Trust. FCRT--Foursight Capital Automobile Receivables Trust. CPSART--CPS Auto Receivables Trust. PART--Prestige Auto Receivables. AMCAR--AmeriCredit Automobile Receivables Trust.

Table 7

Historical ratings activity--U.S. auto loan ABS
Period Upgrades Downgrades
2015 177 0
2016 357 0
2017 322 0
2018 335 2
2019 432 5
2020 332 8
2021 579 0
2022 416 6
2023(i) 305 6
Total 3,255 27
(ii)As of Oct. 31. ABS--Asset-backed securities.

Table 8

Historical ratings activity--Canadian auto loan ABS
Period Upgrades Downgrades
2021 8 0
2022 3 0
2023(i) 0 0
Total 11 0
(i)As of Oct. 31. ABS--Asset-backed securities.

Table 9

Foursight Capital Automobile Receivables Trust 2022-2
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime(i) CNL exp.
2022-2 8.75-9.00 N/A 9.25
(i)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 10

Ford Credit Auto Owner Trust 2022-C
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2022-C 1.10-1.30 N/A 1.00
(i)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 11

CPS Auto Receivables Trust
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2019-A 18.25 11.50(ii) Up to 10.00
2019-B 19.00 11.50(ii) Up to 10.00
2019-C 19.00 12.00(ii) Up to 10.25
2019-D 19.00 12.25(ii) Up to 10.00
2020-B 22.00 12.25(ii) 10.00
2021-A 20.75 12.50(ii) 10.25
2021-C 19.25 16.00(iii) 14.25
2022-A 17.50 19.50(iii) 19.50
2022-C 17.50 N/A 21.00
(i)As of October 2023. (ii)Revised in July 2022. (iii)Revised in May 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 12

CarMax Auto Owner Trust
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2019-4 2.20-2.30 1.35-1.45(ii) Up to 1.40
2020-1 2.15-2.25 1.30-1.40(ii) Up to 1.30
2020-2 2.80-3.00 1.15-1.25(ii) Up to 1.10
2020-3 2.80-3.00 1.30-1.40(ii) 1.25
2020-4 2.80-3.00 1.30-1.40(ii) 1.20
2021-1 2.65-2.85 1.60-1.70(ii) 1.55
2021-2 2.35-2.45 1.60-1.70(ii) 1.50
2021-3 2.15-2.25 2.15-2.25(ii) 2.20
2021-4 2.15-2.25 2.35(iii) 2.35
2022-1 2.15-2.25 2.30(iii) 2.35
2022-2 2.15-2.25 2.45(iii) 2.60
2022-3 2.20-2.30 N/A 2.70
2022-4 2.20-2.30 N/A 2.70
(i)As of October 2023. (ii)Revised in September 2022. (iii)Revised in May 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 13

Carvana Auto Receivables Trust 2022-P3
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2022-P3 2.75 (2.50-3.00) N/A 2.75
(i)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 14

USAA Auto Owner Trust 2022-A
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2022-A 0.60-0.80 N/A 0.55
(i)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 15

Prestige Auto Receivables Trust 2022-1
Series Original lifetime CNL exp. Previous lifetime CNL exp. Revised lifetime CNL exp.(i)
2022-1 16.00 N/A 24.00-25.00
(i)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Table 16

AmeriCredit Automobile Receivables Trust
Series Original lifetime CNL exp. Previous lifetime CNL exp.(i) Revised lifetime CNL exp.(ii)
2019-3 9.75-10.25 5.50 Up to 4.50
2020-2 12.00-12.50 5.00 4.00
2021-1 10.75-11.25 5.50 4.25
2021-2 10.75-11.25 6.50 5.50
2022-2 10.00 N/A 8.50
((i)Revised in December 2022. ii)As of October 2023. CNL exp.--Cumulative net loss expectation. N/A--Not applicable.

Appendix I: Prime And Subprime Index Composition

How do you define prime auto loan ABS?

We generally categorize prime auto loan ABS transactions as those backed by loan pools with initial ECNLs of 3.25% or less and average FICO scores of 700 or higher. We include CarMax and Carvana in this segment.

How do you define subprime auto loan ABS?

We generally categorize subprime auto loan ABS transactions as those backed by loan pools with initial ECNLs of at least 7.5%, average FICO scores of less than 620, and APRs that exceed 14.0%.

How do you calculate the monthly net loss rate?

The monthly net loss rate is annualized. It equals each transaction's net loss rate weighted by the transaction's ending pool balance for the current month over the aggregate ending pool balance of all transactions included in the index.

We only allow a transaction to enter the prime or subprime composite index starting in its fourth month outstanding. Transactions usually have zero or low losses during their first three months, which would dilute the composite data.

How do you calculate the monthly recovery rate?

We calculate recoveries by taking the recovery amount reported (which typically includes all recoveries, including disposition proceeds, post-disposition proceeds, and any other reported recoveries) over the gross loss amount for the current month. We then weight each transaction's recovery percentage by the transaction's ending pool balance for the current month over the aggregate ending pool balance of all transactions included in the index.

We only allow a transaction to enter the prime or subprime index starting in its fourth month outstanding. During a transaction's first three months, unusually high or low recoveries are reported, leading to a spike in the composite figures.

How do you calculate the monthly 60-plus-day delinquency rate?

We calculate delinquencies by taking each transaction's 60-plus-day delinquency amount over the ending pool balance for the current month. Then we weight each transaction's 60-plus-day delinquency percentage by the transaction's ending pool balance for the current month over the aggregate ending pool balance of all transactions included in the composite.

We only allow a transaction to enter the composite starting in its fourth month outstanding. During the transaction's first three months, zero or fewer delinquencies are reported, which would dilute the composite figures.

What is the Auto Loan Static Index (ALSI)?

Our ALSI monitors the credit performance of securitizations that were originated in the same year on a weighted average basis. The number of months displayed for each vintage is generally determined by the last month that all securitizations for that time period have a data point. We calculate the prime and subprime ALSI CNLs by taking the weighted average of the CNLs of the transactions that were completed in the same time period (generally a year). Each transaction's CNL is weighted by its initial pool balance over the aggregate initial pool balance of all transactions included in the index for that period. In the subprime ALSI, transactions from Byrider Finance LLC (doing business as CarNow Acceptance Corp.), Credit Acceptance Corp., and DriveTime Automotive Group Inc. are excluded because they do not have the typical indirect auto loan business model.

Which transactions are included in the prime, subprime, and modified subprime composites and indices?

For a list of the transactions included in our prime and subprime composites and indices, see "U.S. Auto Loan ABS Tracker: June 2023 Performance," published Aug. 15, 2023. However, note that we subsequently added transactions that have since closed.

Related Research

This report does not constitute a rating action.

Primary Credit Analyst:Amy S Martin, New York + 1 (212) 438 2538;
amy.martin@spglobal.com
Secondary Contacts:Jennie P Lam, New York + 1 (212) 438 2524;
jennie.lam@spglobal.com
Steve D Martinez, New York + 1 (212) 438 2881;
steve.martinez@spglobal.com
Sanjay Narine, CFA, Toronto + 1 (416) 507 2548;
sanjay.narine@spglobal.com
Research Contributor:Kapil Sharma, CRISIL Global Analytical Center, an S&P affiliate, Mumbai

No content (including ratings, credit-related analyses and data, valuations, model, software, or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced, or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor’s Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and any third-party providers, as well as their directors, officers, shareholders, employees, or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness, or availability of the Content. S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an “as is” basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT’S FUNCTIONING WILL BE UNINTERRUPTED, OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages.

Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact. S&P’s opinions, analyses, and rating acknowledgment decisions (described below) are not recommendations to purchase, hold, or sell any securities or to make any investment decisions, and do not address the suitability of any security. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment, and experience of the user, its management, employees, advisors, and/or clients when making investment and other business decisions. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. Rating-related publications may be published for a variety of reasons that are not necessarily dependent on action by rating committees, including, but not limited to, the publication of a periodic update on a credit rating and related analyses.

To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw, or suspend such acknowledgement at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal, or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof.

S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain nonpublic information received in connection with each analytical process.

S&P may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, www.spglobal.com/ratings (free of charge), and www.ratingsdirect.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at www.spglobal.com/usratingsfees.

 

Create a free account to unlock the article.

Gain access to exclusive research, events and more.

Already have an account?    Sign in