Credit Acceptance Auto Loan Trust

Transcription

1 Presale: Credit Acceptance Auto Loan Trust This presale report is based on information as of Feb. 7, 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 Preliminary amount (mil. $) Legal final maturity date A AAA (sf) Senior Fixed Feb B AA (sf) Subordinate Fixed April 15, 2027 C A (sf) Subordinate Fixed 75.4 June 15, 2027 (i)ratings are preliminary and subject to change at any time. Profile Expected closing date Feb. 22, Collateral Originator and servicer Seller Nonrecourse loans to dealers secured by subprime automobile installment sales contracts (dealer advances) and subprime automobile installment sales contracts (purchased loans). Credit Acceptance Corp. (BB/Stable/--), a Michigan corporation. Credit Acceptance Funding LLC , a Delaware limited liability company whose sole member is Credit Acceptance Corp. Issuer Credit Acceptance Auto Loan Trust Indenture trustee, trust collateral agent, and backup servicer Owner trustee Initial purchasers Wells Fargo Bank N.A. (AA-/Negative/A-1+). U.S. Bank Trust N.A. Wells Fargo Securities LLC, BMO Capital Markets Corp., Fifth Third Securities Inc., and Credit Suisse Securities (USA) LLC. Primary Credit Analyst: Timothy J Moran, CFA, FRM, New York (1) ; timothy.moran@spglobal.com Secondary Contact: Jenna Cilento, New York (1) ; jenna.cilento@spglobal.com See complete contact list on last page(s) FEBRUARY 7,

2 Credit Enhancement Summary CAALT CAALT Preliminary ratings and amounts (mil. $) Preliminary amount Preliminary amount Class A - 'AAA (sf)' Class B - 'AA (sf)' Class C - 'A (sf)' Total debt Revolving period (mos.) Largest permitted dealer concentration (as a % of NBV) NBV (i.e., balance at cut-off date of dealer advances and purchased loans) (mil. $) Forecast collections (i.e., principal and interest) (mil. $) Installments sales contracts (i.e., principal balance of contracts securing the NBV) (mil. $)(i) , Forecast collections to NBV Relative to NBV Relative to forecast collections Relative to installment sales contracts Relative to NBV Relative to forecast collections Relative to installment sales contracts Overcollateralization (mil. $) Subordination (%) Class A Class B Class C N/A N/A N/A N/A N/A N/A Initial overcollateralization (%) Reserve account(ii) Total initial credit enhancement (%) Class A Class B Class C (i)performing installments sales contracts (i.e., nondefaulted principal balance of contracts securing the NBV) are $813.1 million for series and $586.0 million for series (ii)the reserve account amount is 2% of the rated note balance, nondeclining. CAALT--Credit Acceptance Auto Loan Trust. NBV--Net book value. Rationale The preliminary ratings assigned to Credit Acceptance Auto Loan Trust 's (CAALT 's) $500.0 million asset-backed notes reflect: The availability of approximately 60.5%, 53.1%, and 47.6% credit support for the class A, B, and C notes, respectively, which is based on stressed cash flow scenarios, including excess spread, in respect of the installment FEBRUARY 7,

3 sales contracts. These provide coverage of more than 2.50x, 2.25x, and 1.75x our 19.75%-20.25% expected cumulative net loss range, as a percentage of the performing retail installment sales contracts, for the class A, B, and C notes, respectively. The credit support levels are commensurate with the assigned preliminary 'AAA (sf)', 'AA (sf)', and 'A (sf)' ratings on the class A, B, and C notes, respectively (for more information, see the S&P Global Ratings' Auto Loan Expected Loss and Cash Flow Modeling sections). Our expectation that under a moderate ('BBB') stress scenario, we do not expect our ratings on the notes to decline from our preliminary ratings, all else being equal. Our ratings stability criteria describe the outer bound of credit deterioration over three years as three rating categories in the case of 'AAA', 'AA', and 'A' rated securities (see "Methodology: Credit Stability Criteria," published May 3, 2010). The credit enhancement in the form of subordination, overcollateralization (O/C), a reserve account, and excess spread (see the Credit Enhancement Summary table above). The timely interest and principal payments made under the stressed cash flow modeling scenarios, which we believe are consistent with the assigned preliminary ratings. Credit Acceptance Corp.'s (CAC's) extensive securitization performance history dating back to 2003 and performance history on dealer advances dating back to CAC's long corporate track record and history of consistent profitability. The transaction's payment and legal structures, including a full turbo of the notes following a 24-month revolving period; amortization triggers tied to originator and pool performance; and dealer concentration limits, which, if violated, end the revolving period early and cause all collections to be paid full turbo to the notes sequentially. Changes From The Series Transaction The credit enhancement changes from the series transaction as a percentage of the dealer and purchased loan advances' initial net book value (NBV) include: The total class A credit enhancement increased to 49.93% from 48.21%, while class B's decreased slightly to 33.67% from 33.86%. The class A subordination increased to 28.32% from 26.60%; The class B subordination decreased slightly to 12.06% from 12.25%; and The collateral composition changes from the series transaction pertaining to the dealer advances include: The dollar amount of dealer advances securitized increased to $437.5 million from $309.5 million and decreased relative to the NBV of the pool to approximately 70% from 71%; The number of dealers increased to 1,268 from 1,000; The number of dealer advances increased to 1,500 from 1,117; The top dealer concentration decreased to 0.95% from 1.13% (the maximum dealer concentration remains at 1.50% of the NBV per dealer); The top 10 dealer concentrations decreased to 7.43% from 9.42%; The top state concentration remained in Michigan but increased to 14.61% from 8.12%; and The top 10 state concentrations increased to 61.01% from 60.53%. The collateral composition changes from the series transaction pertaining to the purchased loans' aggregate purchase price include: The dollar amount of purchased loan advances securitized increased to $187.5 million (30.00% of total collateral) FEBRUARY 7,

4 from $128.1 million (29.27% of total collateral); The number of purchased loans increased to 19,663 from 14,539; The top purchased loan advance state concentration decreased to 7.67% (Michigan) from 9.09% (Ohio); and The top 10 purchased loan advance state concentrations decreased to 53.57% from 55.23%. Regarding the advances as a whole, the spread between the series pool's weighted average forecast collection rate and weighted average advance rate remained flat to that of the series pool at 20.3% (see table 1). This spread provides the deal with a margin of safety if the collateral pool experiences a lower collection rate than the forecast. If this spread declines below 16.5%, an automatic trigger immediately ends the revolving period and begins the amortization period. Table 1 shows each securitization pool's spread between its weighted average forecast collection rate and its weighted average advance rate at loan origination. Table 1 Spread Between WA Forecast Collection Rate And WA Advance Rate At Origination(i) CAALT Deal Pool WA forecast collection rate (%) Pool WA advance rate (%) Spread (%) (i)these figures are percentages of the initial origination balances. For series , for example, the forecast collections amount in the credit enhancement summary is as of the Dec. 31, 2017, cut-off date. WA--Weighted average. CAALT--Credit Acceptance Auto Loan Trust. Transaction Overview CAALT is CAC's first securitization in 2018, its 19th stand-alone securitization since 2008, and its 24th S&P Global Ratings credit-rated term securitization. The first distribution will be made on March 15, 2018, and subsequent distributions will be paid on the 15th day of each month or the next succeeding business day. The class A, B, and C notes total $500.0 million. Each class will be paid a fixed interest rate and receive principal sequentially as described in the Payment Structure section below. Chart 1 shows the transaction structure. FEBRUARY 7,

5 In evaluating the credit quality of the securitized assets, we applied our "General Methodology And Assumptions For Rating U.S. Auto Loan Securitizations" criteria, published Jan. 11, Because the installment sales contracts backing the dealer advance portion of the collateral are pledged by the dealers to CAC (as opposed to being sold), we have applied dealer default assumptions to reflect the risk (regardless of how remote) of an automatic stay of these contracts' proceeds in the event of a dealer bankruptcy. (See the S&P Global Ratings' Auto Loan Expected Loss section below for further discussion of our auto loan and dealer default assumptions.) Transaction Structure The series transaction incorporates the following structural features: A 24-month revolving period, during which CAC will add collateral, as needed, to maintain the requisite enhancement levels. A revolving period that may terminate before it is scheduled to end if certain trigger events occur, at which point, FEBRUARY 7,

6 early amortization would begin. A full turbo sequential payment structure across the rated notes at the start of the amortization period, which is scheduled to begin at the close of business on the February 2020 distribution date; however, it is subject to early amortization if a trigger event occurs. An initial O/C of 20% of the NBV that should build quickly during the amortization period because of the full turbo structure. A reserve account that will be funded with an initial deposit of 1.6% of the initial NBV. The reserve account is nondeclining throughout the deal's life. Approximately 70% of the initial NBV consists of dealer advances that were originated under CAC's portfolio program. The remaining 30% consists of the purchase price of the purchased loans (for more information on these programs, see the Business Model section below). The installment sales contracts securing the dealer advances are pledged by the dealers, as opposed to being sold, to CAC. Therefore, our analysis also reflects our view of the risk, however remote, that cash flows from certain installment sales contracts could be delayed by an automatic stay if a dealer whose advances are included in the securitized pool becomes bankrupt. We assumed a certain number of bankrupt dealers at each rating category (see the S&P Global Ratings' Auto Loan Expected Loss section below). As noted above, the full turbo structure causes O/C as a percentage of the NBV to grow quickly once a securitized pool begins to amortize. Table 2 shows the O/C levels of those deals for which data are available at various times during the first six months of amortization, as a percentage of the initial NBV. With the exception of series and , all of the deals listed in table 2 have fully paid down. The more recent transactions, series , , , , and , are still in their revolving periods. Table 2 O/C Levels Deal CAALT Status Amortizing Amortizing Paid-off Paid-off Paid-off Paid-off Paid-off Paid-off O/C level during revolving period (%) O/C level during amortization period (%) Month one Month three Month five Month six N/A O/C--Overcollateralization. CAALT--Credit Acceptance Auto Loan Trust. N/A--Not applicable. As a corollary, the swift paydown of the notes is also presented in the note factor analysis in chart 2. FEBRUARY 7,

7 Chart 2 Trigger events Some of the performance triggers that will cause the revolving period to end and the amortization period to begin include: The adjusted collateral amount, namely the dealer advances and purchased loan collateral, and any amounts on deposit in the principal collections account cannot be less than the amount required to ensure at least 20% O/C for two or more business days. The component of the adjusted collateral amount that consists of a deposit in the principal collections account cannot exceed 5% of the adjusted collateral amount for two or more business days. Cumulative collections for any three consecutive periods must equal at least 90% of the cumulative forecast collections. The spread between the pool's weighted average forecast collection rate and weighted average advance rate must be at least 16.5%. As stated above, the series pool's spread is 20.3%. In rating this transaction, we will review the legal matters that we believe are relevant to our analysis, as outlined in our criteria. FEBRUARY 7,

8 Business Model Incorporated in 1972, CAC targets auto dealerships throughout the U.S. that write installment sales contracts for the subprime and deep subprime auto market. The company has two lending programs: the portfolio and the loan purchase programs. Portfolio program/dealer advances The portfolio program is the older of the two programs and the company's central focus. Under this program, CAC makes loans directly to dealers who are members of its dealer network. These dealer advances, which are securitized, are made in connection with the dealers' origination of installment sales contracts that obligors enter into, to purchase vehicles on the dealers' lots. The installment sales contracts secure the dealer advances and are pledged by the dealers to CAC. The company services those contracts and transfers its secured rights in, and the cash flows from, the contracts to the securitization trust via the seller. The installment sales contract cash flows are used to repay the dealer advances. The portfolio program currently accounts for approximately 70% (as a percentage of NBV) of the series initial pool. Generally, in the portfolio program, each payment received under the installment sales contracts is applied in the following priority: Pay CAC a finance charge that is 20% of the monthly collections for servicing the contracts. Repay the advance CAC made to the dealer. When the advance is paid in full, pay any remaining amounts to the dealer. Loan purchase program In March 2005, CAC introduced the newest version of its loan purchase program, whereby it purchases contracts directly from dealers. This program provides incremental volume and accommodates dealers that do not want to participate in the portfolio program, subject to eligibility requirements. Each monthly payment under the installment sales contract belongs to CAC to cover the contract's purchase price that it paid to the dealer and to pay its finance charge. This program has grown as a percentage of the company's business over the last several quarters. Increasing penetration among large predominantly franchised dealerships, which tend to prefer the purchase loan program, has contributed to an increase in the average amount paid relative to larger, newer, more expensive vehicles purchased over longer terms of up to 72 months. Treatment of finance charges CAC's finance charge is one of the assets transferred to the securitization trust and belongs to the trust. It is used to cover the servicing and trustee fees, with the balance being available to the noteholders. The remaining monthly collections are available for interest, principal, and turboed principal payments to the noteholders after the revolving period. FEBRUARY 7,

9 Collection forecasts CAC's forecast of its expected principal and interest collections on the installment sales contracts originated in relation to advances made under its portfolio program or purchased under its purchased loan program, as well as its ability to actualize that forecast, is central to the business. Using a proprietary statistical model, the company makes a collection forecast for each loan in its portfolio and updates this forecast monthly as it gradually actualizes collections. Table 3 shows the company's original collection forecast on its installment sales contract portfolio by annual vintage, beginning in The initial forecasts are also compared with updated forecasts as of Sept. 30, 2017, and Dec. 31, In addition, the table shows the variance between the initial and most recent forecasts, together with the percentage of the latter that has been realized to date. The forecasts' accuracy increases as the loans season. Furthermore, there was a positive variance between 2008 and The company benefitted from less intense competition and was able to reduce its advance rates in 2008 and The latter vintage proved particularly strong from a collections point of view because the collection rate rose to 79.5% from the initial forecast of 71.9%. By contrast, the 2014 through 2016 vintages currently show negative variance related primarily to heightened competition. Adjusting initial collection forecasts downward in response to the observed decline in forecast collection rates experienced in 2016 may well have adversely affected unit volumes, given the resulting lower advance rates. The 2015 vintage shows the widest negative variance, with the current forecast adjusted down to 65.5% from the initial forecast of 67.7%. Lastly, table 3 shows that the updated forecasts of the vintages are now close to full actualization, rendering these forecasts as essentially reality. Taken as a whole, the table indicates the accuracy with which the company is able to forecast the collections that it expects to make for the installment sales contracts it originates. This is an important consideration because a forecast accuracy of less than 90% for three consecutive months will cause the securitization to begin amortizing. Table 3 Consumer Loan Forecast Collection Rates Initial forecast (%) As of Dec. 31, 2017 (%) As of Sept. 30, 2017 (%) Variance between initial and latest forecast (%) % of forecast realized (0.1) (2.2) (0.6) Although underlying contract defaults can be high, CAC's losses are almost negligible because from 2008 to 2017, it has advanced approximately 42%-47% of the contracts' full principal and interest gross amount under the portfolio program and approximately 46%-52% under the loan purchase program. It has advanced 42% and 46%, for the FEBRUARY 7,

10 portfolio and loan purchase programs, respectively, for the year ended Dec. 31, The company sets each advance rate based on an acceptable level of profitability relative to its forecast collection rate on the installment sales contracts associated with that advance. Both table 4 and chart 3 show the updated forecast/actual collection rates on all of the installment sales contracts by annual vintage, as well as those vintages' associated advance rates, and the spread between the two. The spread provides a cushion if collection rates are lower than those that CAC originally forecast. The spread is also an important risk management metric for both the company's business as a whole and its securitizations. A spread of less than 16.5% between the collection rate and the advance rate during the revolving period will cause the securitization to amortize. From 2008 through 2017, forecast/actual collection rates averaged approximately 71.7%, while advance rates averaged approximately 45.2%, resulting in an average spread of approximately 26.5%. Chart 3 illustrates the percentage of forecast collections actually realized. Table 4 Collection Rate To Advance Rate Spread Forecast/actual collection rate (%)(i) Advance rates (%) Spread (%) Average (i)as of Dec. 31, FEBRUARY 7,

11 Chart 3 The NBV of an identified pool of advances is securitized in these transactions, along with the company's rights in the installment sales contracts associated with those advances. Chart 4 shows the losses on the advances by annual vintage over the past 11 years. The highest loss occurred in 2015 and is currently forecast to be 1.51%. FEBRUARY 7,

12 Chart 4 To help mitigate its risk under the portfolio program, CAC cross-collateralizes the loan repayments made to dealers by aggregating them into pools of approximately 100 loans. Each pool represents the sum of advances made to a dealer and must be repaid from the aggregate collections on the installment sales contracts backing each pool. As stated above, CAC receives 20% of all collections on the installment sales contracts as its finance charge. The 80% balance of monthly collections is first used to pay the advances in each pool. Once all of the advances in a dealer's pool have been repaid, 80% of any subsequent collections are remitted to the dealer as back-end profit. This helps align the interests of CAC with those of each dealer. In addition, not only are a dealer's pool of loans cross-collateralized, all of its pools are themselves cross-collateralized if a dealer has more than one pool. Additionally, a dealer cannot account for more than 1.5% of the securitized dealer advances, nor may the aggregate of the 20 largest dealer concentrations account for more than 20.0% of the securitized dealer advances to mitigate dealer bankruptcy risk. The highest dealer concentration in the series pool at cut-off is approximately 0.95%, and the top 20 dealers account for approximately 12.76% of the securitized dealer advances. The company verifies numerous aspects of a loan application including the income and employment status of each of its obligors. FEBRUARY 7,

13 Similar to other finance companies, CAC has received notices, subpoenas, and civil investigative demands over the last few years from various regulatory agencies such as the U.S Department of Justice, Federal Trade Commission, Consumer Financial Protection Bureau, and various state attorneys general requesting information and documents related to its origination, underwriting, servicing, and securitization of auto loans, as well as information on the company's policies and procedures around allowing car dealers to use GPS starter interrupt devices on vehicles. Payment Structure The distributions will be made from available funds according to the priority shown in table 5. Table 5 Payment Waterfall Priority Payment 1 The servicing fee (6% of monthly collections), the backup servicing fee (10% of monthly collections if the backup servicer has become the servicer or $4,000 if the backup servicer has not become the servicer), and trustee fees and expenses ($2,000). Our cash flow analysis incorporated a 10% servicing fee. 2 Class A, B, and C note interest, paid sequentially. 3 Restore the reserve account to its required level. 4 During the revolving period, purchase additional collateral in an amount sufficient to maintain the required overcollateralization. 5 During the amortizing period, use all available funds sequentially to reduce the class A note principal balance to zero, then to reduce the class B note principal balance to zero, and then to reduce the class C note principal balance to zero. 6 Any unpaid amounts owed to the backup servicer, owner trustee, indenture trustee, and trust collateral agent. 7 Any remainder to the certificateholder. If an event of default occurs under the indenture, the notes will become immediately due and payable, whether automatically, as in the case of the seller or issuer's voluntary or involuntary bankruptcy or by a vote of a majority by principal amount of the holders of the senior-most outstanding notes. If an indenture event of default occurs and the notes have been accelerated, the indenture trustee may exercise certain remedies, such as the liquidation or sale of the trust estate, albeit only in certain circumstances. First, if the event of default is due to a failure to pay principal on the notes at final maturity or the seller's or issuer's voluntary or involuntary bankruptcy, the trust estate may be liquidated without further warning. Second, for any other indenture event of default, the trust estate may be sold or liquidated if all of the noteholders consent, the sale or liquidation proceeds are sufficient to ensure that all of the noteholders are paid in full, or if the indenture trustee determines that the trust estate will not continue to provide sufficient funds to pay principal and interest on the notes as they would have become due had they not been accelerated. If the notes are accelerated in the case of a failure to pay or bankruptcy event, distributions will be made from available funds according to the priority shown in table 6. In the case of any other indenture event of default, distributions will be made from available funds as described in table 5. However, the fees and expenses in item 1 will not be limited, and no amounts will be distributed to the reserve account per item 3. FEBRUARY 7,

14 Table 6 Event Of Default Payment Waterfall Priority Payment 1 The servicing fee; any indemnification amounts owed to the backup servicer; trustee fees, indemnification amounts, and expenses; and unpaid transition expenses due to any successor servicer, paid pari passu. 2 Class A note interest. 3 Class A note principal until the class A note balance has been reduced to zero. 4 Class B note interest. 5 Class B note principal until the class B note balance has been reduced to zero. 6 Class C note interest. 7 Class C note principal until the class C note balance has been reduced to zero. 8 Any remainder to the certificateholder. Pool Analysis As of the Dec. 31, 2017, cut-off date, the series collateral pool contained approximately $ million of advances secured by approximately $1.04 billion principal installment sales contracts. The installment sales contracts' characteristics are shown in table 7. Typical of the company's securitizations, the pool at cut-off contains approximately $ million of principal outstanding related to defaulted contracts to which we did not give any credit in our cash flow runs. The company's collection forecast considers each loan's status, whether performing or defaulted. Table 7 Collateral Comparison Receivables balance (mil. $) No. of receivables Avg. loan balance ($) Weighted avg. APR (%) Weighted avg. original term (mos.) Weighted avg. remaining term (mos.) Weighted avg. seasoning (mos.) CAALT , ,217 63, ,777 80,497 72,729 90,816 69,427 9,614 10,602 8,858 9,165 9,576 8,421 8, FEBRUARY 7,

15 Table 7 Collateral Comparison (cont.) Weighted avg. FICO score Weighted avg. FICO unavailable (%) Original term of mos. (%) Original term of mos. (%) % of used vehicles Year of origination (%) 2010 or older N/A N/A Top five models (%) Chevrolet=17.67 Chevrolet=16.96 Chevrolet=16.96 Chevrolet=16.30 Chevrolet=16.33 Chevrolet=16.42 Chevrolet=18.08 Distribution by model year (%) 2005 or older Ford=14.01 Ford=14.20 Ford=13.58 Ford=14.17 Ford=14.26 Ford=14.30 Ford=14.86 Nissan=9.92 Nissan=10.15 Nissan=9.15 Dodge=8.89 Nissan=8.77 Dodge=8.96 Dodge=9.23 Dodge=8.20 Toyota=9.09 Dodge=8.76 Nissan=8.82 Dodge =8.74 Nissan=8.44 Nissan=7.01 Toyota=8.04 Dodge=7.96 Toyota=7.64 Toyota=7.77 Toyota=7.40 Toyota=6.38 Toyota= N/A Top five customer state concentrations (%) MI=12.41 MI=7.92 MI=10.94 MI=10.01 MI=10.98 MI=10.43 MI=10.68 OH=7.02 OH=7.52 OH=7.57 NY=6.41 OH=8.12 NY=7.84 NY=8.72 TX=5.53 TX=6.64 NY=6.89 TX=5.92 TX=7.84 OH=5.72 OH=6.18 NY=5.48 NY=6.43 TX=6.38 AL=5.67 NY=7.56 PA=5.72 TX=5.55 AL=4.49 MD=5.19 MD=4.61 OH=5.41 MD=5.80 TN=5.67 AL=5.19 Total FEBRUARY 7,

16 Table 7 Collateral Comparison (cont.) APR--Annual percentage rate. S&P Global Ratings' Auto Loan Expected Loss: 19.75%-20.25% To derive the base-case loss for the CAALT transaction, we considered: Net loss static pool data relating to the company's paid-off and outstanding amortizing securitizations; Net loss static pool data for the proposed securitization; The extensive origination net loss static pool data provided by the company; The collateral characteristics of the pool of installment sales contracts securing the securitized pool of advances; The pool's seasoning; Portfolio recovery data; and Our forward-looking view of the auto sector and macroeconomy. Chart 5 shows the origination cumulative net loss static pools by annual vintage from 2002 through With no seasoning credit, they suggest a weighted average net loss expectation of approximately 22%. Chart 5 FEBRUARY 7,

17 The losses on the series , , , , and paid-off deals are summarized in table 8. Because these deals are paid off, it is possible to isolate the dollar amount of losses on the underlying installment contracts attributable to the amortization period. This dollar amount can be measured against the installment contracts' principal balance as of the beginning of the amortization period plus the cumulative collateral added throughout the amortization period to derive a total loss percentage. Table 8 Cumulative Net Loss Summary--Paid-Off Deals Contract balance at beginning of amortization period ($) Total contract balance added during amortization period ($) Total amortization period losses ($) Total amortization period losses (%) ,659,252 86,285,705 45,171, ,695,432 50,562,798 31,591, ,462,203 82,523,113 48,144, ,279, ,808,060 73,204, ,004,825,414 97,826,925 86,728, The losses on the amortizing deals are summarized in table 9. The series and deals are the only deals currently in their amortization periods. After 12 months of amortizing, series is currently at a 20.97% note factor; therefore, it is relatively close to its clean-up call of 10% of the initial note balance. The current cumulative net losses sustained during the amortization period to date are 7.34%, measured as a percentage of the installment contracts' principal balance as of the beginning of the amortization period plus the cumulative collateral added to any existing open pools throughout the amortization period to date. After five months of amortizing, series is currently at a 65.87% note factor. The current cumulative net losses sustained during the amortization period to date are 3.84%. We projected losses on series and using the loss curve derived from the loss timing curves of the paid-off series , , , , and We project these pools to incur total cumulative net losses during their amortization periods of 7.74%, and 9.41%, respectively. These percentages are measured against the installment contracts' principal balance as of the beginning of the amortization period plus the cumulative collateral added through the amortization period. Table 9 Cumulative Net Loss Summary--Amortizing Deals Months amortizing Note factor (%) Contract balance at beginning of amortization period ($) Total contract balance added during amortization period to date ($) Total amortization period losses to date ($) Total amortization period losses to date (%) Total amortization period loss projections (%) ,806,955 57,285,524 58,057, ,042,056 21,542,780 29,508, Series , , , , and are still in their respective revolving periods, have not breached any early amortization triggers, and continue to maintain their required O/C and reserve account amounts. Chart 6 shows the cumulative recoveries on the company's portfolio as a whole. This indicates cumulative recoveries FEBRUARY 7,

18 of approximately 35%, though the average is closer to 25% over the first three years. Chart 6 To derive the 20.00% expected loss, we gave some credit to seasoning but considered that the company will be funding additional collateral during the transaction's two-year revolving period and that this additional collateral could be of a somewhat weaker credit profile. We believe this risk is limited due to the triggers regarding forecast collection accuracy and the minimum 16.50% spread requirement between the collection and advance rates, both referenced in the Business Model section above. If either trigger is breached, the securitization will automatically amortize. Using a 2.50x multiple at 'AAA', 2.25x at 'AA', and 1.75x at 'A' equates to stress losses of 50%, 45%, and 35%, respectively. As previously discussed, we assume the bankruptcy of a certain number of dealers at each rating category under both the largest dealer default amount method and the hybrid default amount method. We further assume that each bankrupt dealer represents the 1.5% per dealer maximum concentration amount permitted in the transaction documents subject to a 95% severity. Our rating-specific cumulative net loss assumption is equal to the greater of the two results, which, in the case of the series transaction, is engendered by the hybrid default amount method. Largest dealer default amount method Under the largest dealer default amount method, at the 'AAA' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top 24 dealer concentrations, aggregating to 34.2% (i.e., 1.5% per dealer FEBRUARY 7,

19 maximum concentration for each of these dealers and allowing for 5.0% recoveries). At the 'AA' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top 20 dealer concentrations, aggregating to 28.5%, albeit the top 20 dealers cannot form more than 20.0% of the advances. At the 'A' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top eight dealer concentrations, aggregating to 11.4%. These concentrations are, in turn, applied to the performing contracts securing the advances made under CAC's portfolio program, which comprise approximately 71.5% of total performing contracts. This means that the 'AAA', 'AA', and 'A' required credit enhancement under the largest dealer default amount method is 38.7%, 33.2%, and 18.1%, respectively. Table 10 Largest Dealer Default Amount Method Stress level AAA Dealer advance portion (71.5%) Number of dealers 24 Maximum concentration per dealer (%) 1.50 Dealer concentration penalty (%) % severity on the dealer concentration penalty (%) Purchased loan portion (28.5%) Base-case loss (%) Multiple (x) 2.50 Total purchased loan portion loss (%) Weighted average loss Dealer advance portion (71.5%) Purchased loan portion (28.5%) Combined loss level (%) Hybrid method Under the hybrid default amount method, at the 'AAA' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top five dealer concentrations, aggregating to 7.5% plus a 50.0% loss on the remaining collateral balance, according to our auto loan criteria (see table 11). At the 'AA' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top four dealer concentrations, aggregating to 6.0% plus a 45.0% loss on the remaining collateral balance. At the 'A' level, the transaction is required to cover a 95.0% loss severity on the collateral associated with the top three dealer concentrations, aggregating to 4.5% plus a 35.0% loss on the remaining collateral balance. This means that the 'AAA', 'AA', and 'A' required credit enhancement under the hybrid method and for this transaction is approximately 52.4%, 47.2%, and 36.9%, respectively. Cash Flow Modeling We modeled the performing portion of the underlying installment sales contracts to simulate the rated stress scenarios appropriate for the assigned preliminary ratings (see table 11). FEBRUARY 7,

20 Table 11 Hybrid Method Cash Flow Assumptions And Results Class A B C Preliminary rating AAA (sf) AA (sf) A (sf) Hybrid method Dealer advance portion (71.5%) Number of dealers Maximum concentration per dealer (%) Dealer concentration penalty (%) % severity on the dealer concentration penalty (%) Base-case loss (%) Base-case loss adjusted for dealer concentration penalty (%)(i) Multiple (x) Adjusted stress loss level (%) Total dealer advance portion loss (%) Purchased loan portion (28.5%) Base-case loss (%) Multiple (x) Total purchased loan portion loss (%) Weighted average loss (%) Dealer advance portion (71.5%) Purchased loan portion (28.5%) Combined loss level (%) Cumulative net loss timing (mos.) 12 12/24 12/24 Cumulative net loss (%) /100 90/100 ABS voluntary prepayments (%) Recoveries (%) Recovery lag (mos.) Servicing fee (%)(ii) 10% of monthly collections 10% of monthly collections 10% of monthly collections Approximate break-even levels (%)(iii) (i)for 'AAA', 18.5% equals 20% base-case cumulative net loss multiplied by 1 minus 7.5% dealer concentration penalty. For 'AA', 18.8% equals 20% base-case cumulative net loss multiplied by 1 minus the 6.0% dealer concentration penalty. For 'A', 19.1% equals 20% base-case cumulative net loss multiplied by 1 minus the 4.5% dealer penalty. (ii)the servicing fee is 6% of monthly collections but was run at the 10% rate that applies if the backup servicer is appointed to take over servicing. (iii)the maximum cumulative net losses on the pool that the transaction can withstand without triggering a payment default on the relevant note classes. ABS--Absolute prepayment speed. The break-even results show that the preliminary rated class A, B, and C notes are credit-enhanced to the degree necessary to withstand stressed net loss levels that are consistent with the assigned preliminary ratings. FEBRUARY 7,

21 Sensitivity Analysis In addition to running break-even cash flows, we ran a sensitivity analysis to see how higher-than-expected losses could affect our preliminary ratings on the notes (see table 12 and chart 7). Table 12 Scenario Analysis Summary Stress level BBB Hybrid method Dealer advance portion (71.5%) Number of dealers 2 Maximum concentration per dealer (%) 1.50 Dealer concentration penalty (%) % severity on the dealer concentration penalty (%) 2.85 Base-case loss (%) Base-case loss adjusted for dealer concentration penalty (%)(i) Multiple (x) 1.40 Adjusted stress loss level (%) Total dealer advance portion loss (%) Purchased loan portion (28.5%) Base-case loss (%) Multiple (x) 1.40 Total purchased loan portion loss (%) Weighted average loss (%) Dealer advance portion (71.5%) Purchased loan portion (28.5%) 7.98 Combined loss level (%) Cumulative net loss timing (mos.) 12/24/36 Cumulative net loss (%) 61/95/100 ABS voluntary prepayments (%) 0.60 Recoveries (%) Recovery lag (mos.) 4 Servicing fee (%) 10% of monthly collections Preliminary 'AAA (sf)' rated notes 2.68x at month one and reaches 5.79x in month 12 when it is paid in full. Preliminary 'AA (sf)' rated notes 2.26x at month one, 4.32x in month 12, and continues to grow thereafter. Preliminary 'A (sf)' rated notes 1.94x at month one, 3.51x in month 12, and continues to grow thereafter. (i)19.4% equals 20% base-case cumulative net loss multiplied by 1 minus 3.0% dealer concentration penalty. ABS--Absolute prepayment speed. FEBRUARY 7,

22 Chart 7 Scenario analysis: 29.44% cumulative net loss results In our 'BBB' scenario, the transaction's credit enhancement is sufficient to cover a 100% loss severity on the collateral associated with the top two dealer concentrations, aggregating to 2.85% as part of a 29.44% total loss on the collateral. Under this stressed loss scenario, the transaction reaches its highest O/C level as a percentage of the initial performing pool balance in month four of the amortization period, which is approximately 42.87% of the initial performing pool balance, and reduces to approximately 25.20% of the initial performing pool balance by the time the notes are fully paid off in month 26. Consistent with the full turbo structure of the amortization period, excess cash flow is released once all of the notes have paid off starting in month 26. The reserve account is never drawn and is released in month 27. Interest is paid on a timely basis for all classes. The class A, B, and C notes are paid full principal in months 14, 20, and 26 of the amortization period, respectively. Our sensitivity analysis allows us to simulate a moderate, or 'BBB', loss scenario to determine the degree to which the preliminary ratings are susceptible to a negative rating action. Given the results noted in table 12, all else being equal, we do not expect our ratings on the class A, B, and C notes to decline from our preliminary 'AAA (sf)', 'AA (sf)', and 'A (sf)' ratings, respectively. FEBRUARY 7,

23 Our ratings stability criteria describe the outer bounds of credit deterioration over three years as three rating categories in the case of 'AAA', and 'AA', and 'A' rated securities. For more information, see "Methodology: Credit Stability Criteria," published May 3, Credit Acceptance Performance: Surveillance Update We currently maintain ratings on seven outstanding CAALT transactions. Five of the seven transactions are still in their revolving period: series issued in May 2016, series issued in October 2016, series issued in February 2017, series issued in June 2017, and series issued in October Series and are currently in their amortization period, which began in January 2017 and August 2017 respectively. We expect series , , , , and to begin their scheduled amortization periods in May 2018, October 2018, February 2019, June 2019, and October 2019, respectively. We recently raised our ratings on the series class B and C notes to 'AAA (sf)' from 'AA (sf)' and 'A (sf)', respectively. We also affirmed the 'AAA (sf)' ratings on the outstanding series class B and C notes. These rating actions reflect our views regarding future collateral performance, as well as each transaction's structure and credit enhancement, among other factors. We will continue to monitor the performance of each transaction and take rating actions as we deem appropriate. 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, 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, FEBRUARY 7,

24 Related Research Two Ratings Raised, Three Ratings Affirmed On Two Credit Acceptance Auto Loan Trust Transactions, Jan. 18, 2018 Credit Acceptance Corp., Aug. 29, 2017 Global Structured Finance Scenario And Sensitivity Analysis 2016: The Effects Of The Top Five Macroeconomic Factors, Dec. 16, 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 primary analyst would like to thank Joe Fang for his analytical contributions to this presale report. Analytical Team Primary Credit Analyst: Timothy J Moran, CFA, FRM, New York (1) ; timothy.moran@spglobal.com Secondary Contact: Jenna Cilento, New York (1) ; jenna.cilento@spglobal.com FEBRUARY 7,

25 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 acknowledgment 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 non-public 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 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. FEBRUARY 7,