Exeter Automobile Receivables Trust

Transcription

1 Presale: Exeter Automobile Receivables Trust This presale report is based on information as of April 13, 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 AA (sf) Senior Fixed June 15, 2021 B A (sf) Subordinate Fixed May 16, 2022 C BBB+ (sf) Subordinate Fixed April 17, 2023 D BB (sf) Subordinate Fixed Feb. 15, 2024 (i)the rating on each class of securities is preliminary and subject to change at any time. (ii)the interest rates and actual sizes of these tranches will be determined on the pricing date. Profile Expected closing date April 26, Collateral Originator, servicer, and sponsor Depositor Structuring lead manager Indenture trustee, backup servicer, and custodian Owner trustee Subprime auto loan receivables. Exeter Finance Corp. EFCAR LLC, a Delaware limited liability company and wholly owned special-purpose subsidiary of Exeter Finance Corp. Citigroup Global Markets Inc. Wells Fargo Bank N.A. (AA-/Negative/A-1+). Wilmington Trust N.A. Primary Credit Analyst: Steve D Martinez, New York (1) ; steve.martinez@spglobal.com Secondary Contact: Ines A Beato, New York (1) ; ines.beato@spglobal.com See complete contact list on last page(s) APRIL 13,

2 Credit Enhancement Summary Subordination (% of the initial receivables)(i) Class A Class B Class C (series , , and rated 'BBB (sf)' and series , and rated 'BBB+ (sf)') Class D Overcollateralization Initial (% of the initial receivables) Target (% of the current receivables)(ii) Floor (% of the 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 (series and rated 'BBB (sf)' and series , series , and rated 'BBB+ (sf)') Class D Excess spread per year (estimated %)(iii) (i)principal on the preliminary rated notes will be paid sequentially. (ii)target overcollateralization steps up to 17.25% upon a cumulative net loss trigger breach for the transaction. (iii)includes the 3.0% servicing fee. --Exeter Automobile Receivables Trust. Rationale The preliminary ratings assigned to Exeter Automobile Receivables Trust 's ( 's) $ million automobile receivables-backed notes series reflect: The availability of approximately 53.5%, 43.5%, 37.1%, and 28.1% credit support for the class A, B, C, and D notes, respectively, based on stressed cash flow scenarios (including excess spread), which provide coverage of approximately 2.50x, 2.05x, 1.72x, and 1.27x our expected cumulative net loss (CNL; see the Cash Flow Modeling section). These break-even scenarios withstand cumulative gross losses of approximately 85.5%, 69.6%, 59.3%, and 44.9%, respectively. The timely interest and principal payments that we believe will be made to the preliminary rated notes under stressed cash flow modeling scenarios that we believe are appropriate for the assigned preliminary ratings. The expectation that under a moderate ('BBB') stress scenario (1.55x our expected loss level), all else being equal, our ratings on the class A notes will remain at the assigned preliminary ratings ('AA (sf)') and our ratings on the B and C notes will remain within one rating category of the assigned preliminary 'A (sf)' and 'BBB+ (sf)' ratings, respectively. The class D notes will remain within two rating categories of the assigned preliminary 'BB (sf)' rating during the first year, but will eventually default under the 'BBB' stress scenario, after having received 70% of their APRIL 13,

3 principal. These rating movements are within the limits specified by our credit stability criteria (see "Methodology: Credit Stability Criteria," published May 3, 2010). The collateral characteristics of the subprime automobile loans securitized in this transaction. The transaction's payment, credit enhancement, and legal structures. Changes From The Transaction The structural changes from the transaction include: Total hard credit enhancement increased for the class A and C notes, to 46.50% and 20.00% from 44.35% and 18.45%, respectively. It decreased for the class B and D notes, to 28.65% and 6.75% from 30.10% and 7.50%, respectively. The initial overcollateralization (O/C) decreased to 4.75% of the initial pool balance from 5.50%, and the target O/C decreased to 12.75% of the current pool balance from 16.00%. The subordination for the class A and C notes increased to 39.75% from 36.85% and to 13.25% from 10.95%, respectively; whereas subordination for the class B notes decreased to 21.90% from 22.60%. Estimated excess spread increased to 14.35% per year from 14.08% per year. The step-up O/C target decreased to 17.25% of the current pool balance from 20.50% upon a breach of the transaction's CNL trigger. The collateral changes from the transaction include: The portion of loans backed by new vehicles decreased to 19.92% from 24.51%. The percentage of loans with a FICO 600 and greater decreased to 26.13% from 28.23%. The weighted average FICO decreased slightly to 574 from 577. The weighted average seasoning decreased to approximately 1.7 months from 4.3 months. The representation in the transaction documents that all contracts in the pool have made at least one payment was removed. The portion of loans with a term of months decreased slightly to 83.36% from 84.54%. The portion of loans with no FICO score increased to 8.17% from 7.46%. The portion of loans originated by the CarMax Operating Cos. (CarMax) channel increased slightly to approximately 20% from 18%. Since early 2016, loans originated through the CarMax channel have performed better than those originated by Exeter's core origination channel. The weighted average loan-to-value (LTV) ratio was generally the same at % compared to %. The payment-to-income ratio increased slightly to 10.61% from 10.40%. In our view, the collateral characteristics for the series pool are marginally weaker compared to those for the series pool. We increased our expected CNL range for this transaction to 20.10%-21.10% from 19.75%-20.75% for series primarily because the representation that all contracts in the pool have made at least one payment was removed. Key Rating Considerations Based on our review of Exeter Finance Corp.'s (Exeter's) operations, we considered the following strengths in rating the transaction: APRIL 13,

4 Multiyear committed funding: The company has a multiyear line of credit, which is provided by five banks (Wells Fargo Bank N.A., Deutsche Bank AG, Citibank N.A., Goldman Sachs & Co., and Barclays Bank PLC) totaling $1.65 billion. The Blackstone Group L.P.'s (Blackstone's) significant investment in Exeter and its ongoing ability and willingness to support the business. Blackstone, Exeter's majority owner, has invested $472 million in total capital in the company and has an incremental $125 million reserved from Blackstone Capital Partners VI to support future growth. Origination history: The company's quarterly static pool data includes more than eight years of performance. It also includes static pool performance for its CarMax origination channel. Securitization history: The company has a five year securitization track record. Over this time, many of its transactions have received upgrades and none of its S&P Global Ratings-rated issuances have been downgraded. However, and breached their performance triggers earlier this year. The company implemented automated proprietary credit scoring across its whole operation when it consolidated its business into two central locations in mid This has promoted stronger operating controls over underwriting and pricing. Backup servicer: Wells Fargo Bank N.A. (AA-/Negative/A-1+), as the backup servicer, indenture trustee, and the loan files' custodian, completed the data mapping for Exeter's servicing system, which we believe will facilitate a smoother servicing transfer, if necessary. Lockbox account: The collections from the loans in the pool are remitted to a lockbox account at Wells Fargo Bank N.A. and then transferred within two days to a collection account also at Wells Fargo Bank N.A. The lockbox account and the collection account are in the indenture trustee's name. Despite the strengths outlined above, we believe that the following limitations, taken as a whole, weigh against assigning a preliminary rating above 'AA (sf)' to the transaction: The company has remained unprofitable on an annual basis since However, Exeter reported a meaningful pre-tax profit for In our view, the company's rapid growth from 2011 through 2014 came at the expense of credit quality and profitability. Furthermore, following a change in leadership in late 2014, the company hastily consolidated its branch network into two central locations and this caused losses on its second-quarter 2015 through fourth-quarter 2015 originations to rise to record levels in full balance charge offs. However, with the subsequent change in senior leadership in early 2016, new operational processes have been implemented to improve asset credit quality. Over the past three years, the company has experienced a high level of senior leadership turnover. The company has had three CEOs in three years. Despite the recent leadership changes, we view the February 2016 appointments of Jason Grubb and Brad Martin as CEO and COO, respectively, as positives given the significant amount of subprime auto finance experience they had at Santander Consumer USA working with Steve Zemaitis, Exeter's president, who originally joined Exeter in 2014 as chief risk officer. The subprime segment of the auto finance market continues to face intense competition. This will continue to put pressure on independent companies' profit margins, especially if funding costs rise and recovery rates on repossessed vehicles continue to decline. Exeter has received subpoenas and Civil Investigative Demands (CIDs) from various regulatory bodies requesting documents related to its origination, underwriting, and securitization of auto loans. Most recently, it received CIDs from the Consumer Financial Protection Bureau seeking information and documents related to its servicing activities. We understand from the company that it is cooperating fully with these requests. At this time, we do not believe these inquiries affect our ratings on the company's asset-backed securities. However, we will continue to monitor ongoing developments and will re-evaluate as we deem appropriate. APRIL 13,

5 Transaction Overview is Exeter's second term asset-backed transaction of Exeter Automobile Trust will issue $ million of securities backed by retail installment sales contracts for new and used automobiles and light-duty trucks. The transaction is structured as a true sale of the receivables to EFCAR LLC (the depositor) from Exeter (the sponsor and servicer) 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). APRIL 13,

6 Transaction Structure The transaction incorporates the following structural features: A sequential-pay mechanism among the notes that increases credit enhancement as the pool amortizes. Initial O/C of 4.75% of the initial pool balance that will build to a target of 12.75% of the current pool balance by using excess spread after net losses to pay principal on the outstanding notes. The O/C floor is set at 0.50% of the initial pool balance. CNL triggers that will increase the O/C target to 17.25% from 12.75% of the outstanding pool balance. The triggers will be tested every three months, starting in month three. Once breached, the triggers are curable on the following test date after three calendar months if the CNL rate is below the threshold. A nonamortizing reserve account that will equal 2.00% of the initial pool balance and will be fully funded at closing. Obligors will make payments directly into a lockbox account at Wells Fargo Bank N.A. in the indenture trustee's name, which mitigates co-mingling risk. Payment Structure The class A through D notes will total $ million and have a fixed interest rate. Interest and principal are scheduled to be paid on each monthly distribution date on the 15th day of each month, or the next business day, beginning May 15, Payment distribution before an event of default Before an event of default, distributions will be made from available funds, according to a specific payment priority on each payment date (see table 1). Table 1 Payment Waterfall Priority Payment 1 To the servicer, the servicing fee (3%), any supplemental servicing fees, any reimbursements for mistaken deposits, and any other related amounts. To any successor servicer, the transition fees up to the specified cap. 2 To the indenture trustee, custodian, owner trustee, and backup servicer, any accrued and unpaid fees, expenses, and indemnities, in each case subject to a maximum specified annual limit. 3 Class A note interest. 4 Principal to the extent necessary to reduce the principal balance of the class A notes to the pool balance (the class A principal parity amount). 5 The remaining principal balance of the class A notes on the final scheduled distribution date. 6 Class B note interest. 7 Principal to the extent necessary, after paying items 4 and 5 above, to reduce the combined principal balance of the class A and B notes to the pool balance (the class B principal parity amount). 8 The remaining principal balance of the class B notes on the final scheduled distribution date. 9 Class C note interest. 10 Principal to the extent necessary, after paying items 4, 5, 7, and 8 above, to reduce the combined principal balance of the class A, B, and C notes to the pool balance (the class C principal parity amount). 11 The remaining principal balance of the class C notes on the final scheduled distribution date. 12 Class D note interest. APRIL 13,

7 Table 1 Payment Waterfall (cont.) Priority Payment 13 Principal to the extent necessary, after paying items 4, 5, 7, 8, 10, and 11 above, to reduce the combined principal balance of the class A, B, C, and D notes to the pool balance (the class D principal parity amount). 14 The remaining principal balance of the class D notes on the final scheduled distribution date. 15 To the reserve account, the amount necessary to achieve the specified reserve account amount. 16 Principal to achieve the specified overcollateralization amount (the principal payment amount). 17 To the indenture trustee, owner trustee, backup servicer, and any successor servicer, any fees, expenses, and indemnities that exceed the related cap or annual limitation. 18 All remaining amounts to the certificateholders. Events of default Any of the following will constitute an event of default: A default in the interest payment on the senior-most class of notes that remains uncured for five days. A default in the principal payment on any note's final scheduled distribution date. The issuer fails to observe or perform any of its material covenants or agreements, or its representations or warranties are materially incorrect and not cured for 45 days (up to 90 days in certain cases). The issuer becomes insolvent. The third item above requires notice by the indenture trustee or the holders of at least 25% of the voting rights for the notes outstanding. Payment distribution after an event of default On each payment date, following an event of default related to a breach of a covenant, agreement, representation, or warranty and the acceleration of the notes, available funds will be distributed as described in table 1. However, the payment in item 16 will include all available funds until the total note balance has been reduced to zero. In addition, the fees, expenses, and indemnities in items 1 and 2 will not be limited. On each payment date following an event of default (other than an event of default solely from a covenant, agreement, representation, or warranty breach), and the acceleration of the notes or upon the trust assets' liquidation, available funds will instead be distributed in the priority shown in table 2. Table 2 Payment Waterfall Following An Event Of Default Other Than Covenant, Agreeement, Or Representation And Warranty Breaches Priority Payment 1 To the servicer, custodian, owner trustee, indenture trustee, and backup servicer, certain amounts due and owing to such entities without regard to any caps or annual limitations. 2 Class A note interest. 3 Class A note principal until paid in full. 4 Class B note interest. 5 Class B note principal until paid in full. 6 Class C note interest. 7 Class C note principal until paid in full. 8 Class D note interest. APRIL 13,

8 Table 2 Payment Waterfall Following An Event Of Default Other Than Covenant, Agreeement, Or Representation And Warranty Breaches (cont.) Priority Payment 9 Principal on the class D notes until the notes are paid in full. 10 All remaining amounts to the certificateholders. Servicer termination events Any of the following will constitute a servicing termination: The servicer fails to deliver any required payment to the indenture trustee and it remains unremedied for two business days. The servicer fails to deliver the servicer's certificate by the first business day before the distribution date, or breaches its covenant not to merge, consolidate, or transfer all or substantially all of its assets unless the successor or surviving entity of the merger or consolidation is capable of fulfilling the servicer's duties. The servicer fails to observe or perform in any covenant or agreement materially and adversely affecting the noteholders' rights and it remains unremedied for 45 days. The servicer becomes insolvent. Any materially incorrect servicer representation, warranty, or statement that remains unremedied for 45 days. If a servicer termination occurs, the indenture trustee or a majority of the noteholders holding the senior-most class can terminate the servicer's rights and obligations. Portfolio Performance Exeter's portfolio decreased by approximately 3% to $3.086 billion as of Dec. 31, 2016, from $3.175 billion as of Dec. 30, 2015 (see table 3). Total delinquencies increased to 20.31% as of Dec. 31, 2016, from 18.31% a year earlier. Annualized net charge-offs increased to 10.00% in 2016 from 7.59% in Deterioration in both measures is partly due to the poorly underwritten 2015 loans, made during the company's consolidation, making their way through the portfolio metrics. Additionally, average seasoning has lengthened due to the contraction in the company's portfolio, adding further stress to the statistics. Table 3 Managed Portfolio Principal amount outstanding end of period (mil. $) Year ended Dec , , , , Delinquencies (i) days days plus days APRIL 13,

9 Table 3 Managed Portfolio (cont.) Total delinquencies as a % of principal amount outstanding Net charge-offs(ii) Average principal amount outstanding during the period (mil. $) Annualized net charge-offs as a % of average principal outstanding Year ended Dec , , , , (i)exeter Finance Corp. considers an automobile loan contract delinquent when more than 10% of a contractual payment remains unpaid by the due date. (ii)net charge-offs equal gross charge-offs minus recoveries. Gross charge-offs do not include unearned finance charges and other fees. Recoveries include repossession proceeds received from the sale of repossessed vehicles net of repossession expenses; refunds of unearned premiums from credit life, credit accident, and health insurance, and extended-service contract costs obtained and financed in connection with the vehicle financing; and recoveries from obligors on deficiency balances. Securitization Performance/Surveillance Update We currently maintain ratings on 12 transactions issued between (see table 4 for the series' pool information). On Jan. 12, 2017, we raised our ratings on 19 classes and affirmed seven classes from series , , , , , , , and We also raised our rating on one class and affirmed our rating on another class from series This transaction has since been optionally redeemed in full (see "Twenty Ratings Raised, Eight Affirmed on Nine Exeter Automobile Receivables Trust Transactions," published Jan. 12, 2017). We revised our loss expectations in January 2017 for eight series issued between 2013 and Series , , and are performing better than our previously revised loss expectations. Additionally, these transactions are significantly seasoned and have likely moved beyond their peak loss range; therefore, we revised down the expected CNL range for those transactions. The series , , and the 2015 vintage transactions are all performing worse than our original expectations; therefore, we raised our expected CNL ranges on those transactions due to the higher-than-expected default frequencies and our view of future collateral performance. APRIL 13,

10 Chart 2 Table 4 Outstanding Transaction Pool Information As Of The January 2017 Distribution Month(i) Transaction/series Current month Pool factor Total delinq. 60+ days delinq. Current CNL Initial expected lifetime CNL Revised expected lifetime CNL (as of January 2017) N/A N/A N/A APRIL 13,

11 Table 4 Outstanding Transaction Pool Information As Of The January 2017 Distribution Month(i) (cont.) Transaction/series Current month Pool factor Total delinq. 60+ days delinq. Current CNL Initial expected lifetime CNL Revised expected lifetime CNL (as of January 2017) N/A (i)cumulative net losses reported are net of liquidation proceeds, which include reasonable expenses incurred by the servicer in connection with the collection of such receivable and repossession and disposition of the financed vehicle. --Exeter Automobile Receivables Trust. CNL--Cumulative net loss. N/A--Not applicable. Pool Analysis As of the April 5, 2017, statistical pool comprised approximately $ million in auto loans (see table 5). The pool is seasoned approximately two months, 83.36% of its loans have an original term of months, and as expected for a subprime pool, a large portion of its loans are backed by used vehicles (80.08%). The final pool is not expected to differ materially from the statistical pool. Some of this pool's characteristics appear weaker than those of the 2015 pools, such as the higher percentage of loans in the lowest FICO and proprietary score bands; however, other attributes have improved, such as the weighted average LTV. While our revised the expected cumulative net loss range for the 2015 pools peaks at 23% (2015-3), we believe that the will perform at a lower loss level because the company, in our view, has addressed many of the issues that lead to higher losses on the 2015 pools. These pools, as well as , were impacted by the disruption issues related to the company's consolidation of its branches and loss of experienced personnel. Also, while the CarMax portion of the pool is lower than in the 2015 through pools (these loans had performed better than Exeter's overall originations until 2015), the CarMax 2015 originated loans experienced higher than expected loss levels, and in some cases where on par with Exeter's overall originations. However, the 2016 CarMax originated loans are performing better than those originated in 2015 and are reporting lower losses than Exeter's overall originations again. Table 5 Collateral(i) Pool size (mil. $) No. of receivables CarMax (% of pool) Avg. principal balance ($) APR original term (mos.) (iii) (iii) ,188 25,841 27,645 19,597 21,180 17,000 25, ,833 16,380 17,045 16,595 17,788 18,287 18, APRIL 13,

12 Table 5 Collateral(i) (cont.) remaining term (mos.) seasoning (mos.) Loans with original term of mos. New vehicles Used vehicles mileage LTV credit bureau score ,631 35,003 37,798 39,251 38,437 38,073 35, No score Less than and greater proprietary credit score (ii) Less than and greater Top three state concentrations TX=16.62 TX=13.68 CA=9.97 CA=9.31 CA=10.56 CA=10.88 TX=11.26% APRIL 13,

13 Table 5 Collateral(i) (cont.) Initial ECNL% Revised ECNL% Pool size (mil. $) No. of receivables CarMax (% of pool) Avg. principal balance ($) APR original term (mos.) remaining term (mos.) seasoning (mos.) Loans with original term of mos. New vehicles Used vehicles mileage LTV credit bureau score CA=9.45 CA=9.34 GA=9.21 GA=8.87 TX=9.32 TX=9.96 CA=10.88% GA=8.56 GA=8.27 TX=8.91 TX=8.56 GA=8.90 GA=9.18 GA=8.80% N/A N/A N/A N/A N/A ,226 27,947 29,713 37,278 30,873 24, N/A N/A N/A 18,549 18,637 17,529 18,163 16,871 16, ,558 34,144 36,799 32,177 35,426 35, No score 3.02 N/A N/A N/A N/A N/A Less than APRIL 13,

14 Table 5 Collateral(i) (cont.) and greater proprietary credit score (ii) Less than and greater Top three state concentrations Initial ECNL% Revised ECNL% TX=12.54 TX=12.27 TX=12.15 TX=11.17 TX=9.67 TX=9.06 CA=10.03 CA=10.32 CA=10.55 CA=9.94 FL=9.07 FL=8.80 FL=9.17 FL=8.48 FL=8.58 FL=9.63 CA=8.62 CA= (i)all percentages are of the principal balance. (ii) The wtd. avg. proprietary credit score is derived prior to the funding of the loan. (iii) Series and included the representation that all obligors made at least one payment. --Exeter Automobile Receivables Trust. CarMax--CarMax Operating Cos. APR--Annual percentage rate. LTV--Loan-to-value ratio. ECNL Expected Cumulative Net Loss S&P Global Ratings' Expected Loss: 20.10%-21.10% To derive the transaction's base-case expected loss, we analyzed Exeter securitizations' CNL and cumulative gross loss (CGL) performance, as well as their origination static pool CNL and CGL performance. The company's first two securitizations, series and , paid off with a CNL of 17.20% and 19.08%, respectively. Collateral losses for series through are trending better than the transaction. We now expect these transactions to lose between 16.25% and 18.50%. Series through are trending noticeably worse than previous securitizations. We now expect these transactions to lose between 19.00% and 23.00%. The origination static pool performance projections have increased slightly since (see chart 3). The average projected losses are approximately 20.74%, 19.44%, 17.74%, 20.81%, and 31.70% for the 2011, 2012, 2013, APRIL 13,

15 2014, and 2015 vintages, respectively. The projected losses for the 2011, 2012, 2013, 2014, and 2015 quarterly vintages range from 18.2%-23.7%, 17.4%-21.3%, 17.0%-18.6%, 18.1%-23.1%, and 23.7%-35.5%, respectively. The 2015 vintage projections are higher than we expect. The loss curve used for the projections is based on prior paid-off vintage performance and may be too slow compared to the ultimate loss curve for The 2016 vintages, while early, are performing better than the 2015 vintages. Chart 3 In addition to the above analysis, we compared the pool's collateral characteristics with those of subprime transactions from its peers. Based on our review of the origination static pool performance, our peer comparison of the collateral pool, securitization performance, and our expectation of lower future recoveries, we expect the transaction to experience losses of 20.10%-21.10%. Cash Flow Modeling: Break-Even Cash Flows We modeled the transaction to simulate stress scenarios appropriate for the assigned preliminary ratings. The cash flow results are based on a pool size of $450.0 million. For these stress scenarios, we applied front- and back-loaded loss curves (see table 6). APRIL 13,

16 Table 6 Cash Flow Assumptions And Results Class Front-loaded loss curve A B C D Scenario (preliminary rating) AA (sf) A (sf) BBB+ (sf) BB (sf) Loss timing by months outstanding (12/24/36/48/60) 53/47/0/0/0 35/32/21/10/1 34/31/20/10/5 34/31/20/10/5 Voluntary ABS Recoveries Recovery lag (mos.) Servicing fee Approximate break-even net loss levels (i) Approximate break-even gross loss levels (i) Back-loaded loss curve Scenario (preliminary rating) AA (sf) A (sf) BBB+ (sf) BB (sf) Loss timing by months outstanding (12/24/36/48/60) 45/55/0/0/0 24/38/31/6/0 21/33/27/16/3 19/31/25/15/10 Voluntary ABS Recoveries Recovery lag (mos.) Servicing fee Approximate break-even net loss levels (i) Approximate break-even gross loss levels (i) (i)the maximum cumulative net losses on the pool that the transaction can withstand without a payment default on the relevant classes of notes. ABS--Absolute prepayment speed. We ran cash flow stress scenarios using the assumptions outlined in table 6 above. We also modeled the transaction's CNL trigger (see table 7 for the CNL trigger levels for each payment date). If the CNL rate exceeds the percentage per the trigger thresholds for any measurement date, the transaction target O/C level will increase to 17.25% of the outstanding pool balance from 12.75% of the outstanding pool balance. The triggers cure if the CNLs stay below the trigger thresholds for three consecutive months, which effectively makes the 12.75% O/C target operative once again. In our aforementioned break-even cash flow scenarios, the CNL trigger is breached and provides added enhancement in only the 'BB' rated back-loaded loss curve scenario. The CNL trigger benefits the transaction in months by requiring a higher target O/C; this delays releasing excess spread and thereby, keeps more credit enhancement in the deal for a longer period. In this scenario, the trigger is breached in month 21 and thereafter. In months 21 30, the available excess spread is used to start building to the 17.25% post-trigger target O/C and reaches a 14.74% peak. The available excess spread is absorbed by the high loss levels in month 31 and thereafter. Therefore, the structure cannot fully build to 17.25% from the 12.75% target. Releases occur in months nine 20 before the CNL trigger is breached. Although the CNL trigger is also breached for classes A, B, and C under both the front- and back-loaded loss curves and for class D under the front-loaded loss curve, the available excess spread is absorbed by the high loss levels so the structure cannot build beyond the initial O/C target of 12.75%. In these scenarios, unlike the 'BB' back-loaded curve, APRIL 13,

17 releases do not occur. Table CNL Trigger Levels Collection period ending Trigger level June September December March June September December March June September December March June September December March Exeter Automobile Receivables Trust CNL Cumulative net loss. The break-even results show that the class A, B, and C notes can withstand higher losses under the back-loaded loss curve than under the front-loaded loss curve because excess spread is more stressed under the front-loaded loss curve. Using the 20.10%-21.10% expected CNL, and applying the above stresses in our internal cash flow runs, the break-even results show that under both the front- and back-loaded loss curves, the class A through D notes are enhanced to the degree necessary to withstand stressed net losses that are consistent with our preliminary ratings. Sensitivity Analysis In addition to analyzing break-even cash flows, we conducted a sensitivity analysis to see how the preliminary ratings on all the notes could be affected by losses that are moderately higher than what we currently expect. For these sensitivity scenarios, we also applied both front- and back-loaded loss curves (see table 8 and charts 4 and 5). Under a back-loaded loss curve in a 'BBB' moderate stress scenario, the CNL trigger is breached in month 18 and thereafter. In months 18 21, the available excess spread is used to build to the post-trigger target O/C (i.e., beyond 12.75%) and reaches a 13.27% peak. In month 22 and thereafter, the available excess spread is again absorbed by the high loss levels, preventing the structure from building to the post-trigger target O/C. Releases occur in months before the CNL trigger is breached. Under a front-loaded loss curve in a 'BBB' moderate stress scenario, although the CNL trigger is breached, the available excess spread is absorbed by the high loss levels so the structure cannot build to the post-trigger target O/C, APRIL 13,

18 and releases do not occur. Table 8 Sensitivity Analysis Summary: Moderate 'BBB' Stress (1.55x Base Case) Front-loaded loss curve CNL level Loss timing by months outstanding (12/24/36/48/60) Voluntary ABS 0.85 Recoveries Recovery lag (mos.) 4 Servicing fee 3.00 Coverage of remaining losses 35/30/20/10/5 Class A ('AA (sf)') Initially 2.06x and then increases steadily, reaching 2.64x by month 12. Class B ('A (sf)') Initially 1.50x and then increases steadily, reaching 1.80x by month 12. Class C ('BBB+ (sf)') Initially 1.23x and then increases steadily, reaching 1.39x by month 12. Class D ('BB (sf)') Back-loaded loss curve CNL level Loss timing by months outstanding (12/24/36/48/60) Voluntary ABS 0.85 Recoveries Recovery lag (mos.) 4 Servicing fee 3.00 Coverage of remaining losses Initially 0.81x and declines gradually over time. A default occurs in month 72 when the pool factor is zero, having repaid 70% of the initial class D note principal balance. 20/30/25/15/10 Class A ('AA (sf)') Initially 2.11x and then increases steadily, reaching 2.38x by month 12. Class B ('A (sf)') Initially 1.55x and then increases steadily, reaching 1.69x by month 12. Class C ('BBB+ (sf)') Initially 1.28x and then increases steadily, reaching 1.35x by month 12. Class D ('BB (sf)') CNL--Cumulative net loss. ABS--Absolute prepayment speed. Initially 0.86x and declines gradually over time. A default occurs in month 72 when the pool factor is zero, having repaid 68% of the initial class D note principal balance. APRIL 13,

19 Chart 4 Chart 5 In our view, under the 1.55x stress scenario all else being equal, our ratings on the class A notes will remain at the assigned preliminary ratings ('AA (sf)') and our ratings on the B and C notes will remain within one rating category of the 'A (sf)' and 'BBB+ (sf)', respectively. The class D rated notes will remain within two rating categories of the assigned preliminary 'BB (sf)' rating during the first year, but will eventually default under the 'BBB' stress scenario, after having received 70% of their principal. These rating movements are within the limits specified by our credit stability criteria (see "Methodology: Credit Stability Criteria," published May 3, 2010). Legal Final Maturity To test the legal final maturity dates set for classes A, B, and C, we determined when the respective notes would be amortized in a zero-loss, zero-prepayment scenario and then added three months to that date. To test the class D legal final maturity date, which is the longest-dated security, we determined the latest maturing loan's distribution date and added nine months to accommodate extensions. Furthermore, in the break-even scenario for each respective rating level, we confirmed that there was sufficient credit enhancement to both cover losses and repay the related notes in full by legal final maturity. Exeter Exeter, a Texas corporation, is a subprime auto finance company that was founded in April In August 2011, three investment funds affiliated with Blackstone acquired an indirect majority interest in Exeter from Navigation Capital Partners and Goldman Sachs & Co. Blackstone has invested approximately $472 million in equity to date and has an incremental $125 million reserved from Blackstone Capital Partners VI to support future growth. As of Dec. 31, 2016, Blackstone, Navigation Capital Partners/Goldman Sachs & Co., and Exeter management own approximately 91.2%, 8.3%, and 0.5% of Exeter, respectively. APRIL 13,

20 Upon receiving the initial capital from Blackstone, Exeter secured an aggregate $600 million multiyear line of credit from a bank consortium now led by Citibank N.A., which has since been renewed, and the commitment is currently up to $1.65 billion. As of Feb. 28, 2017, Exeter had approximately 1,000 employees, two buying centers, and two national servicing centers. The company is licensed in 50 states and has partnerships with approximately 10,000 automobile dealerships. Exeter services a portfolio of approximately 213,000 auto loans with an approximate $3.1 billion aggregate outstanding balance. The company's portfolio has top concentrations in Texas, Florida, and California. The company currently has a partnership with CarMax. Approximately 25.0% of the company's current origination volume is sourced from CarMax. Centralized originations In 2014, Exeter started to transition to a fully centralized and automated originations platform with the introduction and implementation of CALMs. According to management, by the end of second-quarter 2015, the conversion to this automated platform was complete, and credit and funding were centralized in two main offices in Irving, Texas and Clearfield. Branch consolidation In second-quarter 2015, the company closed its branch network. Exeter initially rolled out the branch network to 46 markets to establish local name recognition and a national footprint. The initial originations system did not provide automated decision capability. As a result, the company used local underwriting/credit decisions, which were based on a standard companywide credit policy. The branches began rationalization in As of April 2015, all branches were closed. Management believes the automated underwriting platform and consolidation of its branches will result in tighter controls, efficiency gains, and lower origination costs. Servicing Exeter services its loans through a servicing center in Irving and Clearfield. Delinquencies (up to 60 days past due) are serviced both by the company's collectors and by third-party collection companies that operate in Toronto, Manila, Jamaica, and Colombia. Exeter typically charges off a loan when it is more than 120 days past due at the end of the month if the vehicle is not repossessed or in bankruptcy, or when the company receives sale proceeds from the auction if the vehicle is repossessed. Exeter typically assigns a vehicle to repossession when a loan is 90 days past due and generally liquidates it at auction within 30 days of repossession. Dealer management and pricing In 2016, Exeter created a dealer management department. The department segments dealers by risk grade from high to low based on the dealers' overall historical loss performance. Dealers are proactively deactivated, if appropriate. Since the dealer risk grade implementation, Exeter's origination mix in late 2016 has shifted to consist of more low risk dealerships. The new department also allows Exeter to be more responsive with their dealer partners. APRIL 13,

21 Since Steve Zemaitis, president and chief credit officer, joined Exeter in early 2014, he has focused on improving the pricing of loans that the company originates by introducing price funding scores to achieve a more appropriate risk-adjusted return based on key deal structure elements. These efforts have led to a steady increase in pricing for loans that the company has booked to 21.0% currently from 18.0% in Exeter executive management team Jason Grubb was appointed as the chief executive officer (CEO) of Exeter in February He brings 25 years of leadership to his position at Exeter. His most recent experience includes 11 years at Santander Consumer USA, where he served in leadership roles including president and COO of Originations, in addition to senior vice president of Servicing. Brad Nall was appointed chief financial officer at Exeter in January He joined Exeter in July 2012 and has over 25 years of experience in the consumer finance industry. He has held numerous senior roles during his career within the areas of budgeting and strategic planning, financial reporting, project management, business development, mergers and acquisitions, whole loan sales, and structured finance. Before Exeter, Brad spent 12 years at Citi's auto finance and personal loan businesses in various finance roles. Brad Martin was named chief operating officer of Exeter in February He is a seasoned financial services executive, having served since 2005 in leadership roles with Santander Consumer USA as COO of Servicing and EVP Business Operations. His experience also includes five years at AmeriCredit as a team leader within Originations and Collections. Steve Zemaitis joined Exeter in February 2014 and has almost two decades of experience in the subprime auto finance industry. Prior to joining Exeter, he served as chief credit officer and EVP Pricing & Analytics at Santander Consumer USA Holdings Inc. He led the company's pricing, credit policy, loss forecasting, and scorecard development for retail, lease, and personal unsecured loans. Walter Evans joined Exeter in November 2007 and has held senior legal executive positions at public companies and top-tier law firms for more than 26 years. Before joining Exeter, he was senior vice president and general counsel for ACE Cash Express. Related Criteria And Research Related Criteria 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 - 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, APRIL 13,

22 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 Twenty Ratings Raised, Eight Affirmed On Nine Exeter Automobile Receivables Trust Transactions, Jan. 12, 2017 Global Structured Finance Scenario And Sensitivity Analysis: Understanding The Effects Of Macroeconomic Factors On Credit Quality, Dec. 16, 2016 Presale: Exeter Automobile Receivables Trust , Sept. 29, 2016 Five Exeter Automobile Receivables Trust Deals Reviewed: 10 Ratings Raised, Six Affirmed, April 26, 2016 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 analysts would like to thank Linda Yeh and Kailash Ahirrao for their analytical contribution on this transaction and presale report. Analytical Team Primary Credit Analyst: Steve D Martinez, New York (1) ; steve.martinez@spglobal.com Secondary Contact: Ines A Beato, New York (1) ; ines.beato@spglobal.com APRIL 13,

23 Copyright 2017 by Standard & Poor s Financial Services LLC. All rights reserved. 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. 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, (free of charge), and and (subscription) and (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 STANDARD & POOR'S, S&P and RATINGSDIRECT are registered trademarks of Standard & Poor's Financial Services LLC. APRIL 13,