Drive Auto Receivables Trust

Transcription

1 Presale: Drive Auto Receivables Trust This presale report is based on information as of Feb. 8, The ratings shown are preliminary. This report does not constitute a recommendation to buy, hold, or sell securities. Subsequent information may result in the assignment of final ratings that differ from the preliminary ratings. Preliminary Ratings Class Preliminary rating(i) Type Interest rate(ii) Preliminary amount (mil. $)(ii) Legal final maturity date A-1 A-1+ (sf) Senior Fixed March 15, 2019 A-2-A/A-2-B(iii) AAA (sf) Senior Fixed/floating April 15, 2020 A-3 AAA (sf) Senior Fixed Jan. 15, 2021 B AA (sf) Subordinate Fixed Feb. 15, 2022 C A (sf) Subordinate Fixed March 15, 2023 D BBB (sf) Subordinate Fixed May 15, 2024 E BB+ (sf) Subordinate Fixed June 16, 2025 (i)the rating on each class of securities is preliminary and subject to change at any time. (ii)the tranches' coupons and sizing will be determined on the pricing date. (iii)the class A-2-A notes will be issued as fixed-rate notes, and the class A-2-B notes will be issued as floating-rate notes. The initial principal balance allocation between the class A-2-A and A-2-B notes will be determined on the pricing date. The sponsor doesn't expect the initial principal balance of the class A-2-B notes to exceed $87.13 million. Profile Expected closing date Feb. 21, Collateral Subprime auto loan receivables. Primary Credit Analyst: Amy S Martin, New York (1) ; amy.martin@spglobal.com Secondary Contact: Steve D Martinez, New York (1) ; steve.martinez@spglobal.com See complete contact list on last page(s) FEBRUARY 8,

2 Profile (cont.) Originator, sponsor, servicer, and seller Structuring lead manager Indenture trustee Owner trustee Santander Consumer USA Inc., a subsidiary of Santander Holdings U.S.A. Inc. (BBB/Stable/A-2; formerly known as Sovereign Bancorp). Wells Fargo Securities LLC. Wilmington Trust N.A. (A/Stable/A-1). Wells Fargo Delaware Trust Co. N.A. Credit Enhancement Summary B 2017-A 2016-C 2016-B 2016-A 2015-D 2015-C Subordination (% of the initial receivables)(i) Class A Class B Class C Class D Class E N/A N/A N/A N/A N/A 0.00 Overcollateralization Initial (% of initial receivables) Target (as % of current receivables + as % of initial receivables) Floor (% of initial receivables) Reserve fund (% of the initial receivables) Initial Target Floor Total initial hard credit enhancement (% of the initial receivables) Class A Class B Class C Class D Class E N/A N/A N/A N/A N/A Excess spread per year (estimated %) (iii) (i)principal on the preliminary rated notes will be paid sequentially. Class A refers collectively to the class A-1, A-2, and A-3 notes. (ii)the target overcollateralization amount will increase by 10 percentage points if the CNL trigger is violated. For , this would cause the overcollateralization amount to increase to 39.25% of current receivables plus 2.50% of initial receivables. The CNL triggers are tested monthly and are not curable once breached. (iii)includes the 4.0% servicing fee. --Drive Auto Receivables Trust. N/A--Not applicable FEBRUARY 8,

3 Rationale The preliminary ratings assigned to Drive Auto Receivables Trust 's ( 's) $938 million automobile receivables-backed notes series reflect: The availability of 65.7%, 59.0%, 49.2%, 38.6%, and 35.7% of credit support for the class A (consisting of classes A-1, A-2, and A-3), B, C, D, and E notes, respectively, based on stressed cash flow scenarios (including 100% credit to excess spread), which provide coverage of approximately 2.35x, 2.10x, 1.70x, 1.35x, and 1.27x for our 26.50%-27.50% expected cumulative net loss (CNL; see the Cash Flow Modeling section for more information). These break-even scenarios cover total cumulative gross defaults of 94%, 84%, 70%, 59%, and 55%, respectively. The timely interest and principal payments made under stressed cash flow modeling scenarios are appropriate to the assigned preliminary ratings. The expectation that under a moderate ('BBB') stress scenario (1.35x our expected loss level), all else being equal, our ratings on the class A, B, and C notes will remain at the assigned preliminary 'AAA (sf)', 'AA (sf)', and 'A (sf)' ratings, respectively, and our rating on the class D notes would not decline by more than two rating categories of the assigned preliminary 'BBB (sf)' rating while they are outstanding. The class E notes will remain within two rating categories of the assigned preliminary 'BB+ (sf)' rating during the first year but will eventually default under the front-loaded 'BBB' stress scenario, after having received 73% of its principal, and will be repaid in full under the back-loaded 'BBB' stress. These rating movements are within the limits specified by our credit stability criteria (see "Methodology: Credit Stability Criteria," published May 3, 2010). The originator/servicer's history in the subprime/specialty auto finance business. Our analysis of 10 years of static pool data on Santander Consumer USA Inc.'s (SC's) lending programs. The transaction's payment/credit enhancement and legal structures. Significant Changes From The structural and enhancement changes from the series transaction include: The hard credit enhancement for class A decreased by five basis points (bps), and increased by five bps, 60 bps, 120 bps, and 55 bps for classes B, C, D, and E, respectively. Subordination decreased 60 bps and 50 bps for classes A and B, respectively, and increased five bps and 65 bps for classes C and D, respectively. The decreases in subordination for classes A and B were largely offset by the increase in initial overcollateralization (O/C). Initial O/C increased 55 bps to 19.25% from 18.70%, and target O/C increased slightly to 29.25% of the current pool plus 2.50% of the initial pool from 29.20% of the current pool plus 2.50% of the initial pool. The trigger target O/C also increased slightly to 39.25% of the current pool plus 2.50% of the initial pool from 39.20% of the current pool plus 2.50% of the initial pool. The class E notes will bear an interest rate, which at the time of announcement is expected to approximately 5% p.a., whereas they had a zero percent interest in The estimated excess spread has decreased to 11.52% from 12.38% (pre-pricing). Our preliminary ratings for the class D and E notes is 'BBB (sf)' and 'BB+ (sf)', respectively, compared to 'BBB- (sf)' and 'BB- (sf)', respectively, for series The higher ratings are attributable to 120 bps and 55 bps, respectively, in more credit enhancement at closing, and a reduction in our expected cumulative net loss (ECNL) to a range of 26.5%-27.5% from 27.0%-28.0%. FEBRUARY 8,

4 In our opinion, the significant collateral changes from include: The weighted average FICO of the pool (excluding no FICO loans) increased to approximately 572 from approximately 566. This is the highest weighted average FICO score of the pools to date. The weighted average loss forecasting score (LFS) increased to 476 from 471, and the percentage with an LFS below 450 declined to approximately 34% from 39%. The percentage of longer-term loans (73-75 month) decreased slightly to 8.6% from 9.4%. The percentage of new vehicles increased to 34.6% from 33.6%. The percentage of the pool with no FICO increased to 10.9% from 10.5%. The pool seasoning decreased to 3.7 months from 5.3 months. The Texas and Florida state concentrations returned to more normal levels for pools, increasing to 15.6% and 12.4%, respectively, from 14.6% and 7.1%, given that several months have passed since Hurricanes Harvey and Irma. For the month stratification, the weighted average LFS improved to 485 from 479, and the weighted average FICO increased to 567 from 565. However, the weighted average loan-to-value (LTV) increased slightly to 96.3% from 94.9%. We believe the pool is of slightly higher credit quality than the prior pool. Additionally, with more performance data on 's outstanding securitizations, it now appears that many of them are performing below our original ECNLs. As a result, we've reduced our ECNL range on this transaction to 26.5%-27.5% from 27.0%-28.0%. Key Rating Considerations We considered the following key strengths in rating this transaction: SC, the originator/servicer of the assets, is owned by Santander Holdings U.S.A. Inc. (SHUSA; BBB/Stable/A-2). Further, we consider SHUSA a highly strategic subsidiary of Banco Santander S.A. (A-/Stable/A-2) that would receive support from its parent, which owns 100% of SHUSA, if needed (see "Santander Holdings U.S.A Inc.," published Aug. 1, 2017). SC has a long track record of originating and securitizing subprime auto loans. SC's key officers have many years' experience in the auto business. SC has generally repurchased contracts from its securitized pools that have not made their first two payments, and beginning with the transaction, they included a representation that each contract has or will have made their first two payments. If not, they will repurchase those contracts. With the exception of the 2016-A transaction, all of 's transactions are performing in line to slightly lower than our original ECNLs. Additionally, we have noticed improvement in CNL performance since The overcollateralization target that requires an additive 2.50% of initial receivables is a strong structural feature. This target is effective at trapping excess spread that would otherwise leave the transaction and helps to protect the transaction in the event losses are backloaded. In addition to the strengths outlined above, we considered the following weaknesses: Approximately 82% of the obligors in the pool did not provide proof of income (POI), nearly the same as the 12 outstanding transactions as reported by SC management. While the level of income verification is low for a securitized subprime pool, the company uses various risk identifiers, including credit history, deal structure, dealer risk, and other factors, to determine when POI is warranted. Also, the loans whose obligors had their incomes FEBRUARY 8,

5 verified have higher LTVs (111% versus 108%) than those who did not have their incomes verified. Since the level of income verification has been relatively low for all of the deals, we believe the loss performance and repurchase activity of its securitizations reflect the company's verification strategy. Since July 2014, the sponsor has received civil subpoenas and civil investigative demands from various federal and state agencies--including from the Department of Justice, the SEC, and an executive committee representing attorneys general of 30 states--requesting documents that, among other things, relate to the origination, underwriting, and securitization of auto loans for varying time periods since These types of investigations and proceedings could result in adverse consequences to SC. Indeed, in March 2017, SC entered into settlements (aggregate amount of $26 million) with the attorneys general of Massachusetts and Delaware to settle allegations that it facilitated the origination of certain Massachusetts and Delaware loans that it knew--or should have known--violated applicable state consumer protection laws. The two states alleged that the obligors' income levels had been inflated. The above settlement-covered loans purchased in those states from Since then, the company has expanded its dealer review process. The risk of default from income inflation is often detected during the loan's earliest stages (first two payments). SC has consistently repurchased loans that have failed to make their first two payments, and beginning with the transaction, it has included a representation that each contract has received or will have received its first two payments. If not, SC will repurchase them. None of SC's securitizations have paid-off (the first transaction was completed in March 2015), so we don't have a paid-off loss curve with which to project losses on the outstanding transactions. However, we used paid-off loss curves from Santander's Santander Drive Auto Receivables Trust (SDART) transactions and a derived loss curve from 2015-A to project losses on 's outstanding securitizations. The percentage of loans in 's pools with no FICO is significant at 10.9%. However, it is line with prior pools (10.5% for ), and all of the loans in the pool have an LFS. Transaction Overview In 2015, SC established for deep subprime assets that do not conform to the collateral characteristics of its existing platforms: SDART, Chrysler Capital Auto Receivables Trust (CCART), and Santander Prime Auto Issuance Notes (SPAIN) is the 13th transaction issued from this platform will issue $938 million in securities backed by retail installment sales contracts for new and used automobiles and light-duty trucks. SC originated all of the receivables included in the pool under its standard credit and underwriting procedures. The transaction is structured as a true sale of the receivables to Santander Drive Auto Receivables LLC (the depositor) from SC (the originator, sponsor, servicer, and seller), and then to (the issuer), which has pledged its interest in the receivables to the indenture trustee on the noteholders' behalf. In rating this transaction, S&P Global Ratings will review the legal matters that it believes are relevant to its analysis, as outlined in its criteria. (See chart 1 for the transaction structure.) FEBRUARY 8,

6 Transaction Structure incorporates the following structural features: A sequential-pay mechanism among the notes that results in increased credit enhancement for the rated notes as the pool amortizes. Initial O/C of 19.25% of the initial pool balance that will build to a target that equals 29.25% of the current pool balance plus 2.50% of the initial pool balance by using any excess spread available after covering net losses to pay principal on the outstanding notes. The O/C floor is set at 2.50% of the initial pool balance. A loss trigger that will step up the O/C target to 39.25% of the current pool balance plus 2.50% of the initial pool balance if the collateral performance deteriorates and losses exceed the loss threshold. The triggers will be tested monthly and are not curable once breached. A nonamortizing reserve account that will be 1.00% of the initial pool balance and will be fully funded at closing. FEBRUARY 8,

7 Payment Structure All of the classes will pay a fixed interest rate, except for class A-2-B, which will pay a floating rate. Interest and principal are scheduled to be paid to the rated notes on each monthly distribution date on the 15th day of each month or, if that is not a business day, the next business day, beginning March 15, On each payment date, distributions will be made from available funds according to the payment priority in table 1. Table 1 Payment Waterfall Priority Payment 1 To the indenture and owner trustees, and the asset representations reviewer any accrued and unpaid fees and any reasonable expenses not previously paid by the servicer, at a $300,000 cap per year, in aggregate. 2 To the servicer, the servicing fee (4.0%) and all unpaid servicing fees from previous periods. 3 To the class A noteholders, the accrued class A note interest due for the related interest period (paid pro rata). 4 The first principal allocation (the excess, if any, of the class A note balance over the pool balance). 5 To the class B noteholders, the accrued class B note interest due for the related interest period. 6 The second principal allocation (the excess, if any, of the combined class A and B note balances over the pool balance minus the first principal allocation). 7 To the class C noteholders, the accrued class C note interest due for the related interest period. 8 The third principal allocation (the excess, if any, of the combined class A, B, and C note balances over the pool balance minus the first and second principal allocations). 9 To the class D noteholders, the accrued class D note interest due for the related interest period. 10 The fourth principal allocation (the excess, if any, of the combined class A, B, C, and D note balances over the pool balance minus the first, second, and third principal allocations). 11 To the class E noteholders, the accrued class D note interest due for the related interest period. 12 The fifth principal allocation (the excess, if any, of the combined class A, B, C, D, and E note balances over the pool balance minus the first, second, third, and fourth principal allocations). 13 To the reserve account, any additional amounts required to make the cash on deposit in the reserve account equal to the specified reserve account balance. 14 The regular principal allocation, if any (in this step, the notes are paid down to build to the overcollateralization target). 15 To the certificate distribution account for the residual interestholder, any funds remaining. Events Of Default Under the indenture, the occurrence and continuation of any of the following events constitutes a default: A default in the interest payment on any of the controlling class' notes when due and payable that continues for five or more business days. A default on the principal payment on any note on the final scheduled payment or redemption date. The issuer fails to duly observe or perform its covenants or agreements. Any of the issuer's representations or warranties made under the indenture proves incorrect. The issuer files for bankruptcy. FEBRUARY 8,

8 Payment Priority After An Event Of Default On each payment date after an event of default occurs, available funds will be distributed in the priority shown in table 2. Table 2 Payment Priority After An Event Of Default Priority Payment 1 Any accrued and unpaid fees to the indenture and owner trustees and the asset representations reviewer. 2 The servicing fee (4.0%) and all unpaid servicing fees to the servicer. 3 Interest, pro rata, to the class A noteholders. 4 If an acceleration of the notes occurs after, or as a result of, the event of default detailed in the first, second, or fifth bullet in the Events Of Default section above, pay in the following priority: principal to the class A-1 noteholders until the notes are paid in full; then principal to the class A-2 and A-3 noteholders, pro rata, based on each class' note balance until all class A notes are paid in full; then accrued class B note interest to the class B noteholders; then principal to the class B noteholders until the notes are paid in full; then accrued class C note interest to the class C noteholders; then principal to the class C noteholders until the notes are paid in full; then accrued class D note interest to the class D noteholders; then principal to the class D noteholders until the notes are paid in full; then accrued class E note interest; and then principal to the class E noteholders until the notes are paid in full. 5 If an acceleration of the notes has occurred after, or as a result of, the event of default detailed in bullets three or four in the Events Of Default section above, pay in the following priority: accrued class B note interest to the class B noteholders; then accrued class C note interest to the class C noteholders; then accrued class D note interest to the class D noteholders; then accrued class E note interest to the class E noteholders; then principal to the class A-1 noteholders until the notes are paid in full; then principal to the class A-2 and A-3 noteholders, pro rata, based on each class' note balance until all of the class A notes are paid in full; then principal to the class B noteholders until the notes are paid in full; then principal to the class C noteholders until the notes are paid in full; and then principal to the class D noteholders until the notes are paid in full; and then principal to the class E noteholders until the notes are paid in full. 6 Any remaining funds will be deposited into the certificate distribution account for distribution to or at the direction of the residual interestholder. Managed Portfolio Performance SC's managed portfolio performance is starting to show early signs of stabilization as of Sept. 30, 2017: 31-plus day delinquencies declined to 15.53% from 15.63% a year earlier. While annualized net losses rose 4.5% year over year for the nine months ended Sept. 30, 2017, this increase was a substantial reduction from the 22% increase in losses in 2016 (9.43% compared to 7.75% in 2015). Management has attributed the increase in losses to a greater mix of deep subprime originations in 2015, and, in our view, has responded by reducing origination volume, especially within the higher-risk tiers. SC's managed portfolio declined approximately 4% to $24.7 billion as of Sept. 30, 2017, from $25.7 billion a year earlier. In addition, in its 2017 full-year earnings presentation, it noted that originations below a 640 LFS declined by $486 million for the fourth quarter of 2017 compared to the same period in Table 3 shows the performance for SC's total subprime auto loan portfolio, including contracts securitized under both the SDART and platforms. There is a higher concentration of lower-credit quality receivables in the pool than in the loss and delinquency tables presented below. FEBRUARY 8,

9 Table 3 Managed Portfolio Principal amount outstanding at end of period (mil. $) As of Sept. 30 As of Dec ,668 25,687 25,247 26,498 22,862 21,128 16,206 14,139 14,801 Delinquencies (%) days days plus days and bankrupt accounts Total 31-plus-day delinquencies as a % of the principal amount outstanding(i) Net loss experience Average principal amount outstanding during the period (mil. $) Annualized net losses as a % of the average principal outstanding 25,135 26,250 26,045 25,459 22,499 18,918 15,124 14,325 10, (i)the servicer considers a receivable delinquent when an obligor fails to pay the required minimum portion of the scheduled payment by the due date, as determined per SC's customary servicing practices. For Chrysler Capital receivables originated by the servicer before Jan. 1, 2017, the required minimum payment is 90% of the scheduled payment. For all other receivables the servicer originated or acquired from an unaffiliated third-party originator before Jan. 1, 2017, the required minimum payment is 50% of the scheduled payment. For receivables the servicer originated or acquired from an unaffiliated third-party originator on or after Jan. 1, 2017, the required minimum payment is 90% of the scheduled payment, regardless of which channel originated the receivable. In each case, the period of delinquency is based on the number of days payments are contractually past due. SC--Santander Consumer USA Inc. Pool Analysis As of the Jan. 31, 2018, cut-off date, the collateral pool comprised approximately $1.16 billion in SC-originated auto loans (see tables 4A and 4B for a collateral comparison with the prior securitizations). In our view, the series pool is of better credit quality than series , and substantially improved since series 2016-A (currently the highest-loss pool). For example, for series , the weighted average LTV has declined to 108.3% from 111.8% for 2016-A, the percentage of loans with an LFS of 450 or less has decreased to approximately 34% from 43%, and the weighted average FICO has risen to approximately 572 (the highest FICO of all of the transactions) from 550. Also, the percentage of loans with an original term loan of months has decreased to 8.6% from 9.4% for These longer-term loans have a weighted average LTV of 96.3%, which is below the pool's overall LTV of 108.3%, and have a weighted average LFS of 485, which is higher than the pool's overall LFS of 476 and the highest LFS across all prior pools. FEBRUARY 8,

10 Table 4A Collateral Comparison(i) (v) Issuance date Feb. 21, 2018 Oct. 25, 2017 July 31, 2017 June 28, B March 29, A 2016-C Jan. 31, 2017 Nov. 22, B May 25, 2016 Pool size (mil. $) 1,161.71(ii) 1,543.75(ii) 1,180.45(ii) 1,382.67(ii) 1, , , , SC (% of pool) Avg. original principal balance ($) Weighted avg. APR (%) Weighted avg. original term (mos.) 17,813 18,540 19,150 20,277 18,291 19,702 18,040 18, Seasoning (mos.) New vehicles (%) Used vehicles (%) Weighted avg. LTV ratio (%) Weighted avg. original FICO score(iii) No score or lower or higher Weighted avg. internal credit score(iv) or lower or higher Total % of loans with an original term of mos. Total % of loans with an original term for mos. Weighted avg. original FICO score for mos. Weighted avg. internal credit score for mos FEBRUARY 8,

11 Table 4A Collateral Comparison(i) (cont.) Weighted avg. LTV ratio for mos (v) B 2017-A 2016-C 2016-B Top five state concentrations (%) S&P Global Ratings' expected lifetime CNL (%) S&P Global Ratings' most recent revised expected lifetime CNL (%) Securitization performance Cumulative net loss as of January 2018 distribution date (%) TX=15.62 TX=14.58 TX=16.24 TX=15.70 TX=16.68 TX=15.90 TX=16.60 TX=17.23 FL=12.43 FL=7.10 FL=10.16 FL=11.11 FL=11.18 FL=12.12 FL=12.69 FL=16.09 CA=8.19 CA=7.08 GA=7.94 GA=7.50 CA=8.37 CA=8.90 CA=9.47 CA=10.86 GA=5.91 GA=6.60 CA=7.69 CA=7.25 GA=6.58 GA=6.81 GA=6.41 GA=5.53 MD=3.97 PA=6.08 NY=4.63 NY=4.26 NC=4.17 NC=4.20 NY=3.81 NY= N/A N/A N/A N/A N/A N/A N/A N/A N/A Pool factor (%) N/A Months outstanding N/A (i)all percentages are of the initial gross receivables balance. (ii)beginning with , the pools exclude diesel engine equipped Jeep Grand Cherokee SUVs and Dodge Ram 1500 trucks that are the subject of the May 2017 Justice Department lawsuit against Fiat Chrysler for emission test violations. (iii)excludes receivables of obligors with no FICO scores. (iv)sc uses a proprietary internal credit scoring methodology to underwrite and originate collateral. The score ranges from one to 999, with a score of one indicating a high predicted likelihood of loss and a score of 999 indicating a low predicted likelihood of loss. (v)for Drive , any loan that SC believes was affected by Hurricane Harvey or Irma was removed from the final pool. --Drive Auto Receivables Trust. SC--Santander Consumer USA Inc. APR--Annual percentage rate. CNL--Cumulative net loss. LTV--Loan-to-value. N/A--Not applicable. Table 4B Collateral Comparison(i) A 2015-D 2015-C 2015-B 2015-A Issuance date Feb. 21, 2018 Jan. 27, 2016 Sept. 29, 2015 July 22, 2015 May 28, 2015 March 18, 2015 Pool size (mil. $) 1,161.71(ii) , , SC (% of pool) Avg. original principal balance ($) 17,813 17,378 18,344 18,318 18,879 18,583 Weighted avg. APR (%) Weighted avg. original term (mos.) Seasoning (mos.) New vehicles (%) Used vehicles (%) FEBRUARY 8,

12 Table 4B Collateral Comparison(i) (cont.) A 2015-D 2015-C 2015-B 2015-A Weighted avg. LTV ratio (%) Weighted avg. original FICO score(iii) No score or lower or higher Weighted avg. internal credit score(iv) or lower or higher N/A N/A Total % of loans with an original term of mos. Total % of loans with an original term for mos. Weighted avg. original FICO score for mos. Weighted avg. internal credit score for mos. Weighted avg. LTV ratio for mos. Top five state concentrations (%) S&P Global Ratings' expected lifetime CNL (%) S&P Global Ratings' most recent revised expected lifetime CNL (%) Securitization performance Cumulative net loss as of January 2018 distribution date (%) TX=15.62 CA=12.89 TX=14.63 TX=13.75 TX=20.75 TX=19.81 FL=12.43 TX=11.95 FL=12.14 FL=12.15 FL=13.83 FL=11.84 CA=8.19 FL=10.61 CA=11.16 CA=11.43 CA=9.02 CA=10.18 GA=5.91 GA=6.25 GA=5.91 GA=6.13 GA=4.34 GA=5.57 MD=3.97 NY=4.43 IL=4.25 IL=4.44 NC=3.89 IL= N/A N/A N/A Pool factor (%) N/A FEBRUARY 8,

13 Table 4B Collateral Comparison(i) (cont.) A 2015-D 2015-C 2015-B 2015-A Months outstanding N/A (i)all percentages are of the initial gross receivables balance. (ii)beginning with , the pools have excluded diesel engine equipped Jeep Grand Cherokee SUVs and Dodge Ram 1500 trucks that are the subject of the May 2017 Justice Department lawsuit against Fiat Chrysler for emission test violations. (iii)excludes receivables of obligors with no FICO scores. (iv)sc uses a proprietary internal credit scoring methodology to underwrite and originate collateral. The score ranges from one to 999, with a score of one indicating a high predicted likelihood of loss and a score of 999 indicating a low predicted likelihood of loss. (v)for Drive , any loan that SC believes was affected by Hurricane Harvey or Irma was removed from the final pool. --Drive Auto Receivables Trust. SC--Santander Consumer USA Inc. APR--Annual percentage rate. CNL--Cumulative net loss. LTV--Loan-to-value. N/A--Not applicable. Securitization Performance And Surveillance The outstanding securitizations are generally performing in line with or better than our original ECNLs. Drive's 2015 securitizations have experienced weighted average CNLs of 16.11% through month 28. The 2016 securitizations have experienced a weighted average CNL of 8.63% through month 14, which is slightly higher than the 2015 vintage at the same point (8.39%). This uptick is primarily due to higher losses on the 2016-A securitization (see chart 3). The 2016 transactions are also experiencing weaker recoveries compared to the 2015 transactions (see chart 4). Beginning in 2016, the company started to net out repossession expenses from liquidation proceeds. This, coupled with lower used vehicle values, has negatively affected its recovery rates since early Performance has improved since the 2016-A transaction with the subsequent transactions experiencing CNLs in line to slightly lower than for We believe the improved performance is the result of the improved credit quality of 's securitizations, including lower LTVs. We currently maintain ratings on 12 transactions issued from 2015 through fourth-quarter On June 15, 2017, we reviewed our loss expectations for the series 2015 vintages and revised/maintained them as shown in table 5 below (see "Seven Ratings Raised, Four Affirmed On Four Drive Auto Receivables Trust Transactions," published June 15, 2017). Due to the deleveraging inherent to these structures, we raised our ratings on seven classes across four transactions. Table 5 CNL Expectations Series Current month Pool factor (%) CNL (%) Original lifetime CNL expectations (%) Updated lifetime CNL expectations (%)(i) 2015-A B C D A N/A 2016-B N/A 2016-C N/A 2017-A N/A 2017-B N/A N/A FEBRUARY 8,

14 Table 5 CNL Expectations (cont.) Series Current month Pool factor (%) CNL (%) Original lifetime CNL expectations (%) Updated lifetime CNL expectations (%)(i) N/A N/A (i)for series 2015-A through 2015-D, revised as of June 15, CNL--Cumulative net loss. N/A--Not applicable. For all outstanding classes, the credit support levels have increased as a percentage of the declining collateral balances. In addition, the O/C for each series can step up to a higher target O/C level if certain CNL metrics are breached. In our view, all of the classes have adequate credit enhancement at their current rating levels. Chart 2 FEBRUARY 8,

15 Chart 3 FEBRUARY 8,

16 Chart 4 Peer Analysis Two peers that are roughly in 's deep subprime expected loss range are DriveTime and American Credit Acceptance (ACA). However, when we compared their securitizations, it is clear that the platform performs better, with lower CNLs (see charts 5 and 6). Our current ECNL for the ACA 2015, 2016, and 2017 transactions ranges between 25.0% and 29.5% depending upon the transaction. Our current ECNL for the DriveTime transactions for the same vintages ranges between 27.0% and 30.5%, depending upon the transaction. This peer analysis lent support to our lower ECNL for this transaction. FEBRUARY 8,

17 Chart 5 FEBRUARY 8,

18 Chart 6 S&P Global Ratings' Expected Loss: 26.50%-27.50% We examined static pool data from SC, broken out by the company's LFS and by term, which was further segmented by LFS. The LFS is based on the obligor's credit quality, the loan structure, and other noncredit-bureau attributes. We used the 2009, 2010, and 2011 vintage aggregate paid-off loss curves to project losses for the outstanding SC-originated collateral by internal score band, only projecting vintages with at least 12 months of performance. We then applied the pool composition weights to the projected losses of each LFS band to determine a weighted average loss level for the pool. We also used projected loss levels for the various loan terms further segmented by LFS band to arrive at another weighted average loss proxy for the pool. Based on our review of the origination static pool performance by internal score band and internal score band segmented by term, securitization performance, peer analysis, and our forward-looking view of the economy, we expect to experience CNLs in the 26.5%-27.50% range. FEBRUARY 8,

19 Cash Flow Modeling: Break-Even Cash Flows Cash flow modeling tests the availability and timing of excess spread and stresses the release of enhancement. Excess spread, which can be a critical component of a transaction's overall credit enhancement, can be affected by many factors, such as the absolute level and timing of defaults, prepayment speeds, payment timing lags, and the collateral terms. We modeled the transaction to simulate stress scenarios appropriate for the assigned preliminary ratings (see table 6). Table 6 Cash Flow Assumptions And Results Class Front-loaded loss curve A B C D E Scenario (preliminary rating) AAA (sf) AA (sf) A (sf) BBB (sf) BB+ (sf) Loss timing input by months outstanding (12/24/36/48) (%) Loss timing output by months outstanding (12/24/36/48) (%) 40/40/10/10 40/40/10/10 40/40/10/10 40/40/10/10 40/40/10/10 100/0/0/0 55/45/0/0 42/39/10/9 42/40/9/9 42/40/9/9 Voluntary ABS (%) Recoveries (%) Recovery lag (mos.) Servicing fee (%) Approximate break-even net loss levels (100% credit to excess spread) (%)(i) Approximate break-even gross loss levels (100% credit to excess spread) (%)(i) Back-loaded loss curve Scenario (preliminary rating) AAA (sf) AA (sf) A (sf) BBB (sf) BB+ (sf) Loss timing input by months outstanding (12/24/36/48/60) (%) Loss timing output by months outstanding (12/24/36/48/60) (%) 20/30/25/15/10 20/30/25/15/10 20/30/25/15/10 20/30/25/15/10 20/30/25/15/10 100/0/0/0/0 49/51/0/0/0 30/40/30/0/0 25/33/26/15/0 23/32/25/14/7 Voluntary ABS (%) Recoveries (%) Recovery lag (mos.) Servicing fee (%) Approximate break-even net loss levels (100% credit to excess spread) (%)(i) Approximate break-even gross loss levels (100% credit to excess spread) (%)(i) (i)the maximum cumulative net and gross losses on the pool that the transaction can withstand without a payment default on the relevant classes of notes. ABS--Absolute prepayment speed. FEBRUARY 8,

20 Loss Trigger Table 7 CNL Threshold Changes Month CNL threshold (%) and thereafter CNL--Cumulative net loss. Front-loaded loss break-even analysis no benefit from trigger In the front-loaded break-even scenarios, the trigger was hit early (months five to six for the class A and B runs and months for classes C through E). However, the trigger provided no benefit in any of the runs because O/C started to decline in dollar terms after reaching a maximum of approximately 22%-25% of current receivables depending upon the stressed scenario, which is short of the required O/C target of 29.25% of current receivables plus 2.50% of initial receivables. Back-loaded loss break-even analysis - benefits from O/C feature but not the trigger In the back-loaded scenarios, the trigger provided no benefit (hit in months five, eight, 12, and 36 for classes A, B, C, and D, and never for class E). However, the O/C mechanism proved instrumental in preventing releases in the class D run. Whereas the O/C didn't reach the original target in the class A through C runs, O/C did achieve the 29.25% level in the class D run, but since it was required to trap another 2.50% of initial receivables (which at that specific time, month 15, was about 4% of current receivables), no money could be released. The deal didn't hit the trigger until later at month 36, at which point the deal was not at its pre-trigger required level. In the class E run, the O/C reached the all-in target in month 17 and there were releases from months 17-28, which aggregated to only 1.3% of initial receivables. At month 29, losses started to eat into the O/C and, as a result, the deal fell below the requisite level and there were no further releases. The trigger was never breached in this scenario. Using an expected net loss of 26.50%-27.50% of the initial pool and applying the transaction-appropriate stresses in our internal cash flow runs, the break-even results show that the class A through E notes are enhanced to the degree necessary to withstand stressed net loss levels that are consistent with the assigned preliminary ratings. Sensitivity Analysis In addition to analyzing break-even cash flows, we conducted a sensitivity analysis to determine the potential rating impact of a moderate ('BBB') stress scenario, all else being equal, on our ratings on the notes (see table 8). FEBRUARY 8,

21 Table 8 Sensitivity Analysis Summary: Moderate 'BBB' Stress (1.35x Base Case) Used Floating-Rate Structure Cumulative net loss level (27.00% x 1.35) (%) Loss timing input months outstanding (12/24/36/48) (%) Loss timing output by months outstanding (12/24/36/48) (%) Fast loss curve Slow loss curve 40/40/10/10 20/30/25/15/10 42/40/9/9 24/33/25/14/4 Voluntary ABS (%) Recoveries (%) Recovery lag (mos.) 4 4 Servicing fee (%) Potential rating decline Class A ('AAA (sf)') No rating decline expected No rating decline expected Class B ('AA (sf)') No rating decline expected No rating decline expected Class C ('A (sf)') No rating decline expected No rating decline expected Class D ('BBB (sf)') Decline of up to two rating categories Decline of up to two rating categories Class E ('BB+ (sf)') ABS--Absolute prepayment speed. Class E defaults on principal in this scenario after paying off approximately 73% of the notes Class E is repaid in full. Our sensitivity analysis indicated that in a moderate stress scenario, our ratings on the class A, B and C notes will remain at the assigned preliminary 'AAA (sf)', 'AA (sf)', and 'A (sf)' ratings, respectively, and our rating on the class D notes will remain within two categories of the assigned preliminary 'BBB (sf)' rating over the life of the transaction. Our rating on the class E notes will remain within two rating categories of the assigned preliminary 'BB+ (sf)' rating during the first year, but under the front-loaded loss curve, the bonds will default after receiving 73% of their principal. Under the back-loaded loss curve, the class E notes will be repaid in full. These rating movements are within the limits specified by our credit stability criteria (see "Methodology: Credit Stability Criteria," published May 3, 2010). In the front-loaded moderate stress scenario the trigger was breached in month 12, the O/C peaked at 24.7% of current; as such, the trigger did not provide a benefit. There were no releases in this scenario. In the back-loaded moderate stress scenario, the trigger was never hit, the O/C reached its all-in target in month 17, and there were releases from months FEBRUARY 8,

22 Chart 7 FEBRUARY 8,

23 Chart 8 Money Market Tranche Sizing The proposed money market tranche's (class A-1) legal final maturity date is March 15, To test whether the money market tranche can be repaid by its maturity date, we ran cash flows using assumptions to delay the principal collections. In our cash flow run, we assumed zero defaults and a zero absolute prepayment speed. We also confirmed that approximately 12 months of principal collections would be sufficient to pay off the money market tranche. Legal Final Maturity To test the legal final maturity dates set for the long-dated tranches (classes A-2 through D), we determined when the respective notes would fully amortize in a zero-loss and zero-prepayment scenario and then added three months to the result. For the longest-dated security (class E), we added 12 months to the tenor of the longest receivable in the pool to accommodate extensions on the receivables. Furthermore, in the break-even scenario for each rating level, we confirmed that credit enhancement was sufficient to both cover losses and repay the related notes in full by the legal final maturity date. FEBRUARY 8,

24 Santander Consumer USA Inc. SC was established as an independent entity under the name Drive Financial Services on Aug. 18, Before being acquired by Banco Santander S.A. on Dec. 7, 2006, Drive Financial Services was 64.5% owned by BoS (USA) Drive Inc., a wholly owned subsidiary of HBOS Group PLC, and 35.5% owned by Drive Financial Services' management. Banco Santander S.A. completed its acquisition of Drive Financial Services on Dec. 7, 2006, for $771 million in cash, resulting in a 91.5% ownership of the company. SC has been a part of SHUSA since 2009, when Banco Santander S.A. contributed the business to SHUSA. In December 2011, SC issued stock privately to investors, lowering SHUSA's ownership to 65% from 90%, and at that time, SC was deconsolidated from SHUSA's balance sheet. Then, in January 2014, investors and SHUSA sold a portion of SC stock in an IPO. Following the IPO, SHUSA sold down its ownership to 59% of SC. Even though SHUSA's ownership of the subsidiary was reduced, SC was again fully consolidated on SHUSA's balance sheet after the investor group's stake sale. In our corporate analysis, and in line with the accounting treatment, we have continually viewed SC as if it were wholly owned by SHUSA because we believe that SHUSA bears the risk of this subsidiary and is ultimately responsible for its support. In November 2017 SHUSA increased its ownership in SC to 68.1% by acquiring the approximate 9.6% stake owned by SC's former CEO who resigned in July SC, based in Dallas, originates and services auto loans and leases through a network of more than 17,000 dealers nationwide. Since February 2013, it has also served as Chrysler Group LLC's preferred lender by originating auto loans across the full credit spectrum. In 2017, SC originated approximately $20.2 billion in auto-related receivables, comprising 37% core subprime auto loans, 33% Chrysler Capital retail auto loans, and 30% Chrysler Capital auto leases. As of Dec. 31, 2017, it had $39 billion in assets, $35 billion in finance receivables and leases, and $6.5 billion in shareholder equity. Including servicing for third parties, its average managed consumer portfolio totaled $50.1 billion in The company reported that it earned a net income of $1.2 billion in Excluding the impact of significant items, including tax reform, legal reserves, and a settlement with the former CEO, net income was $627 million. Banco Santander S.A., headquartered in Madrid, is a worldwide banking organization that was established in 1857 and is a European leader in consumer finance with activities in 13 European countries, particularly Spain, Germany, Italy, and Poland. Banco Santander S.A. owns 100% of SHUSA, which now has approximately 59.0% equity interest in SC. Banco Santander S.A. is the largest financial institution by assets in Spain, and the credit rating is related to the sovereign credit ratings on Spain (BBB+/Stable/A-2). On Aug. 28, 2017, Jason Kulas was replaced as CEO for SCUSA by Scott Powell. Mr. Powell had previously been serving as CEO for SHUSA since March 2015, and had served on the SCUSA board since September On Oct. 2, 2017, SC announced that Juan Carlos Alvarez, Corporate Treasurer of SHUSA, would succeed Ismail (Izzy) Dawood as chief financial officer of SC, effective immediately. In addition, Sandra Broderick joined SC as executive vice president, head of operations on Oct. 10, overseeing originations, servicing and default, and other operations functions including administrative oversight of Santander Consumer International, Puerto Rico. FEBRUARY 8,

25 Related Criteria General Criteria: Methodology For Linking Long-Term And Short-Term Ratings, April 7, 2017 Criteria - Structured Finance - General: Ratings Above The Sovereign - Structured Finance: Methodology And Assumptions, Aug. 8, 2016 Criteria - Structured Finance - General: Methodology: Criteria For Global Structured Finance Transactions Subject To A Change In Payment Priorities Or Sale Of Collateral Upon A Nonmonetary EOD, March 2, 2015 Criteria - Structured Finance - General: Criteria Methodology Applied To Fees, Expenses, And Indemnifications, July 12, 2012 General Criteria: Global Investment Criteria For Temporary Investments In Transaction Accounts, May 31, 2012 Criteria - Structured Finance - RMBS: U.S. Interest Rate Assumptions Revised For May 2012 And Thereafter, April 30, 2012 Criteria - Structured Finance - ABS: General Methodology And Assumptions For Rating U.S. Auto Loan Securitizations, Jan. 11, 2011 Criteria - Structured Finance - General: Standard & Poor's Revises Criteria Methodology For Servicer Risk Assessment, May 28, 2009 Legal Criteria: Legal Criteria For U.S. Structured Finance Transactions: Special-Purpose Entities, Oct. 1, 2006 Legal Criteria: Legal Criteria For U.S. Structured Finance Transactions: Appendix III: Revised UCC Article 9 Criteria, Oct. 1, 2006 Legal Criteria: Legal Criteria For U.S. Structured Finance Transactions: Criteria Related To Asset-Backed Securities, Oct. 1, 2006 Legal Criteria: Legal Criteria For U.S. Structured Finance Transactions: Securitizations By Code Transferors, Oct. 1, 2006 Legal Criteria: Legal Criteria For U.S. Structured Finance Transactions: Select Issues Criteria, Oct. 1, 2006 Related Research Santander Holdings U.S.A Inc., Aug. 1, 2017 Seven Ratings Raised, Four Affirmed On Four Drive Auto Receivables Trust Transactions, June 15, 2017 Global Structured Finance Scenario And Sensitivity Analysis 2016: The Effects Of the Top Five Macroeconomic Factors, Dec. 16, 2016 Four Ratings Raised and 20 Ratings Affirmed From six Drive Auto Receivables Trust Deals, Oct. 19, 2016 Banco Santander And Banco Bilbao Vizcaya Argentaria Upgraded On Spain Action; Outlook Stable; Some Banks Affirmed, Oct. 6, 2015 In addition to the criteria specific to this type of security (listed above), the following criteria articles, which are generally applicable to all ratings, may have affected this rating action: "Post-Default Ratings Methodology: When Does Standard & Poor's Raise A Rating From 'D' Or 'SD'?," March 23, 2015; "Global Framework For Assessing Operational Risk In Structured Finance Transactions," Oct. 9, 2014; "Methodology: Timeliness of Payments: Grace Periods, Guarantees, And Use of 'D' And 'SD' Ratings," Oct. 24, 2013; "Counterparty Risk Framework Methodology And Assumptions," June 25, 2013; "Criteria For Assigning 'CCC+', 'CCC', 'CCC-', And 'CC' Ratings," Oct. 1, 2012; "Methodology: Credit Stability Criteria," May 3, 2010; and "Use of CreditWatch And Outlooks," Sept. 14, The primary analyst would like to thank Ishan Shankar and Samantha Wong for their analytical contributions to this FEBRUARY 8,