AutoTrader (AUTOA.L)

Transcription

1 Europe/United Kingdom Equity Research New Media Rating (from NEUTRAL) UNDERPERFORM Price (07 Mar 17, p) Target price (p) (from ) Market Cap ( m) 3,895.5 Enterprise value ( m) 4,110.2 Target price is for 12 months. Research Analysts Joseph Barnet-Lamb joseph.barnet-lamb@credit-suisse.com Sophie Bell sophie.bell@credit-suisse.com Matthew Walker matthew.walker@credit-suisse.com Stephanie MacAulay stephanie.j.macaulay@credit-suisse.com AutoTrader owns and operates the UK's leading automotive market place. The platform, which is 100% digital after exiting print in 2013, owns AutoTrader.co.uk the UK's leading Auto portal sitting between consumers searching for vehicles and those selling them. AutoTrader (AUTOA.L) DOWNGRADE RATING Cycle set to bite shifting down a gear Downgrade to Underperform: We downgrade AutoTrader (AUTOA) to Underperform and reduce our target price to 360p (from 390p) as we believe the current valuation doesn't correctly reflect short-medium term risks around the cycle, with (Average Revenue Per Retailer) ARPR gains likely to become harder to obtain and longer term risks around autonomous vehicles. Cyclical issues approach: We believe the industry is in cyclical flux with timing hard to call but expect new car sales to slow, pressuring retailer industry profitability. We also expect used car prices to fall eventually, causing some smaller retailers' gross profit margins to fall. Underlying price rises and cyclical pressure will squeeze retailers: We construct a proprietary view of the financials of independent UK used car retailers and hypothesise that the cyclical issues coupled with largely underlying price rises could effectively wipe out the operating margins of some less agile smaller retailers causing many of them to fold. This would lead to a further decline in AutoTrader retailer membership. Autonomous threat won't go away: We continue to believe driverless technology could eventually reduce auto transaction volumes. We don t expect the impact of this evolution to be felt in the near term but believe with press and market attention on the issue rising the threat will not go away. Cut estimates by 1-3%: We cut our underlying operating profit forecasts by 1% (FY17) and 3% (FY18) as we lower forecourt numbers (FY19-2.5%). Our EPS forecasts fall 1.4% (FY17) and 3.1% (FY18) as we update our buyback estimates. We are 2% below consensus for FY18. Risks: The key risk to our call is like-for-like used car prices not falling as expected. Valuation: AutoTrader currently trades on 19.3x CY17E EBITDA and 4.5% FCF yield vs Rightmove on 4.2%; we believe this discount isn't great enough given our heightened cyclical concerns at AUTOA. Our TP is DCF derived. Financial and valuation metrics Year 3/16A 3/17E 3/18E 3/19E Revenue ( m) EBITDA ( m) Pre-tax profit adjusted ( m) CS EPS (adj.) (p) Prev. EPS (p) ROIC (%) P/E (adj.) (x) P/E rel. (%) EV/EBITDA (x) Dividend (03/17E, ) 5.07 Net debt/equity (03/17E,%) Dividend yield (03/17E,%) 1.3 Net debt (03/17E, m) BV/share (03/17E, ) -0.0 IC (03/17E, m) Free float (%) 94.0 EV/IC (03/17E, (x) 12.4 Source: Company data, Thomson Reuters, Credit Suisse estimates DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, LEGAL ENTITY DISCLOSURE AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

2 AutoTrader (AUTOA.L) Price (07 Mar 2017): p; Rating: (from NEUTRAL) UNDERPERFORM; Target Price: (from p) p; Analyst: Joseph Barnet-Lamb Income statement ( m) 3/16A 3/17E 3/18E 3/19E Revenue EBITDA Depr. & amort. (11) (9) (8) (8) EBIT Net interest exp. (15) (12) (12) (12) Associates PBT Income taxes (28) (37) (38) (41) Profit after tax Minorities Preferred dividends Associates & other Net profit Other NPAT adjustments (2) (5) (7) (7) Reported net income Cash flow ( m) 3/16A 3/17E 3/18E 3/19E EBIT Net interest (15) (12) (12) (12) Cash taxes paid (16) (37) (38) (41) Change in working capital (2) (4) (3) (4) Other cash and non-cash items Cash flow from operations CAPEX (3) (10) (9) (9) Free cashflow to the firm Acquisitions Divestments Other investment/(outflows) Cash flow from investments (3) (10) (9) (9) Net share issue/(repurchase) 0 (84) (92) (105) Dividends paid 0 (50) (55) (60) Issuance (retirement) of debt (147) (4) (5) (8) Cashflow from financing (173) (139) (153) (173) Changes in net cash/debt Net debt at start Change in net debt (133) (18) (11) (16) Net debt at end Balance sheet ( m) 3/16A 3/17E 3/18E 3/19E Assets Total current assets Total assets Liabilities Total current liabilities Total liabilities Total equity and liabilities Per share 3/16A 3/17E 3/18E 3/19E No. of shares (wtd avg.) (mn) 1, CS EPS (adj.) (p) Dividend (p) Free cash flow per share (p) Key ratios and valuation 3/16A 3/17E 3/18E 3/19E Growth/Margin (%) Sales growth (%) EBIT growth (%) Net income growth (%) EPS growth (%) EBITDA margin (%) EBIT margin (%) Pretax profit margin (%) Net income margin (%) Valuation 3/16A 3/17E 3/18E 3/19E EV/Sales (x) EV/EBITDA (x) EV/EBIT (x) Dividend yield (%) P/E (x) Credit ratios (%) 3/16A 3/17E 3/18E 3/19E Net debt/equity (%) (766.0) (987.7) (1585.8) ( ) Net debt to EBITDA (x) Interest coverage ratio (x) Source: FTI, Company data, Thomson Reuters, Credit Suisse Securities (EUROPE) LTD. Estimates Company Background AutoTrader operates the UK's leading digital marketplace for the Autos sector. The site connects independent and franchise car dealers with its audience of car buyers in return for monthly subscription fees. Blue/Grey Sky Scenario Our Blue Sky Scenario (p) (from ) We assume rising used car sales volumes make retailers more willing to spend more on AUTOA. As such we see ARPR through to 2020 grow at a CAGR of +12% instead of our current +10% (& 10% 20-25). We assume retailer volumes are flat post FY17. We then also assume the industry sees no adverse impact from driverless vehicle and as such put growth in perpetuity beyond 2025 up to +3% in-line with that used for RMV. Yields a warranted price of 530p. Our Grey Sky Scenario (p) (from ) We assume the glut of nearly new vehicles puts downward pressure on used car prices leading dealers to go out of business and leaves remaining retailers less able to afford AUTOA s services. As such we change two things; firstly we use a membership CAGR (17-20) of 5% vs CSe -1.1%, and in the face of driverless tech we lower growth in perpetuity from +1.5% to -1%. Yields a warranted price of 250p Share price performance May- 15 Sep- 15 Jan- 16 May- 16 Sep- 16 Jan- 17 AUTOA.L FTSE 100 IDX The price relative chart measures performance against the FTSE 100 IDX which closed at on 07/03/17 On 07/03/17 the spot exchange rate was.87/eu 1.- Eu.95/US$1 AutoTrader (AUTOA.L) 2

3 Table of contents Key charts 4 Risks approach in the rear view mirror 5 Cyclicality a growing headwind 6 New car sales have peaked...6 Used car sales lag new car sales, so will likely soften...8 Used car prices matter too and they're slowing...9 UK franchise dealers profitability is already being squeezed...10 US trends support our thesis...12 So what does this all mean for AutoTrader?...13 Retailer solutions need to deliver in FY18 18 Cross-sell a dwindling proportion of ARPR rises...18 Underlying price rises can't go on forever...19 Macro concerns could accelerate issues...20 FY17-18 big years for new product launches...23 Autonomous vehicles a long-term threat 24 Forecast changes 26 TP falls to 360p, rating to Underperform 28 Credit Suisse HOLT Linker...30 AutoTrader (AUTOA.L) 3

4 Key charts Figure 1: Annual new car sales vs two-year lagged used car sales Figure 2: Proportion of YoY absolute ARPR rise by lever 15.0% 10.0% 5.0% 0.0% -5.0% -10.0% E 42% 52% 39% 33% 15% 18% 25% 34% 44% 39% 27% 31% 19% 33% 48% -15.0% H H YoY used car sales - 2 yr lag YoY new car sales Price Stock Upsell/Cross Sell Source: SMMT Source: AutoTrader, Credit Suisse research Figure 3: CSe average UK independent used car retailers* spend c10% of used car gross profit on classified advertising and 9% of AutoTrader services but still have 23.2% of gross profit as operating profit Proportion of gross profit spent on: Staff 33.7% Used car COGS 85% Used car gross margin 15% Overheads (incl property) 28.2% Operating profit 23.2% Operating margin: 3.5% AutoTrader 12.0% Other classified advertising 3.0% Source: ASE, SMMT, AutoTrader, Credit Suisse estimates and research *Excludes retailers who do not use any AutoTrader services Figure 4: CS scenario analysis illustrating the impact of a contraction of used car gross profit margin & rising AUTOA ARPR on less agile independent used car retailers Proportion of gross profit spent on: Staff 40.6% Used car COGS 89% Used car gross margin 11% Overheads (incl property) 35.0% Operating profit 0.9% Operating margin: 0.1% AutoTrader 21.8% Other classified advertising 1.7% Source: ASE, SMMT, AutoTrader, Credit Suisse estimates and research AutoTrader (AUTOA.L) 4

5 Risks approach in the rear view mirror We downgrade our rating on AutoTrader to Underperform (from Neutral) and lower our target price to 360p (from 390p) as we believe the current valuation doesn't correctly reflect short-medium term risks around the cycle, with ARPR gains becoming harder to obtain (especially as cyclical issues start to bite) and longer term risks around autonomous vehicles. In this note, we put forward the following key points; 1) Risks from the cycle approach: We believe the industry is in cyclical flux with timing hard to call but ultimately we expect profit pressure on smaller retailers to cause a number of smaller less agile retailers to go out of business. The premise of our view is; a) We believe that sterling depreciation and an influx of nearly new used cars (falling off the leasing cycle) will erode the relative value proposition of new cars in the UK, thus lowering sales and eroding retailer profitability (new car absolute profitability c.15% higher than used car). b) AUTOA will likely partially be protected from this as short-term used car volumes rise, inflating AUTOA's value add to retailers (95% cars on site are used) and pushing up the stock component of AUTOA's ARPR. c) However, we hypothesise that an eventual downward movement in used car prices (due to heightened supply) could further squeeze retailer profitability for some less agile retailers as they get caught with stock on their forecourt. Ultimately we believe this could cause some retailers to go out of business. d) Whilst used car stock from folded retailers will likely move to remaining retailers, fixed per forecourt pricing will likely be foregone, in our view (CSe c.1/3 pricing is on a fixed per forecourt basis, 2/3 is stock based). 2) Retailer Solutions needs to deliver in FY18: We like AutoTrader s innovative pricing strategy and believe the group s dominant position in the UK online auto space gives it significant underlying pricing power. However, we would flag that the proportion of absolute ARPR rises coming from the group s more sustainable cross-sell of wider value added services has consistently fallen in recent periods (52% FY14 to 19% HY17). With UK auto classified advertising largely now online (CSe 86% 2017) we believe u/l price based ARPR rises will become harder and believe the group needs to deliver on a pipeline of product launches in H2 FY17 and into FY18 to reassure investors future ARPR rises can be delivered. We construct a proprietary view of the financials of independent UK used car retailers and hypothesise that cyclical issues coupled with largely u/l price rises could wipe out the operating margins of some less agile retailers causing many of them to fold. We believe a reduction in gross margin to 11% from 15% for smaller, less agile independent dealers, could lead to up to 1k retailers (20% of the UK's <35 car independents) going out of business. Making conservative assumptions around the movement of stock to remaining larger players would yield a 6m or 3% FY17 u/l operating profit headwind. Our FY17 and FY18 absolute forecourt numbers are 1.2% (-160) and 3.2% (-430) below consensus expectations and our FY18 u/l operating profit is 2% below consensus. Whilst this impact may not be very big, leading online classifieds players enjoy high multiples in part due to their visibility, consistent returns and upgrade cycles. Even relatively minor downgrades could cause meaningful de-ratings, in our view. 3) Autonomous threat won't go away: We continue to believe driverless technology could eventually reduce auto transaction volumes. We don t expect the impact of this evolution to be felt in the near term but believe with press and market attention on the issue rising the threat will not go away. See our July 2016 initiation of AutoTrader. 4) Valuation AUTOA currently trades on 19.3x CY17E EBITDA and 4.5% FCF yield vs RMV on 4.2%, we believe this discount isn't great enough given our heightened cyclical concerns at AUTOA. Our TP of 360p is DCF derived and would put the group on 5.0% FCF yield. We see more attractive options elsewhere in the European Internet space citing our top calls ZPG (OP, TP 440p) and JustEat (OP, TP 680p). AutoTrader (AUTOA.L) 5

6 Cyclicality a growing headwind We believe with the UK new and used car markets in a period of cyclical transition so too is AutoTrader; however, the timing of that transition is hard to call. We believe that sterling depreciation and an influx of nearly new used cars (falling off the leasing cycle) will erode the relative value proposition of new cars in the UK, thus lowering sales and eroding retailer profitability (new car absolute profits are c15% higher than used car). AutoTrader will likely partially be protected from this as used car volumes rise somewhat (AutoTrader value add is higher in used than new); however, we hypothesise that an eventual downward movement in used car prices could further squeeze retailer profitability and cause some retailers to go out of business. Whilst used stock will likely move to remaining retailers, the fixed pricing element of those folded forecourts (c.1/3 of ARPR) would likely be foregone, in our view. Whilst there are multiple stages to the above process and timing is far from certain, it is our view that AutoTrader could see an acceleration of forecourt declines over the next 18 months as industry consolidation continues and forecourt closures step up. Our FY17 and FY18 absolute forecourt numbers are 1.2% and 3.2% below consensus expectations, respectively, at 13,244 for FY17 and 12,913 for FY18. New car sales have peaked After five years of improving new car sales, the new car market looks to have peaked. Although sales grew 2.3% in 2016, momentum slowed significantly through the year, particularly after the Brexit vote. Growth fell from 3.2% in 1H16 to 1.2% in 2H16, ending the year with -1.1% in December. SMMT (The Society of Motor Manufacturers & Traders, a UK automotive industry body) forecast new car registrations down 5.0% in Below we discuss why we too (not withstanding strong January numbers) believe that new car sales likely have peaked. Figure 5: UK new car sales growth 20.0% 15.0% 10.0% 5.0% -5.0% - Jan-03 Feb-04 Mar-05 Apr-06 May-07 Jun-08 Jul-09 Aug-10 Sep-11 Oct-12 Nov-13 Dec-14 Jan % -15.0% -20.0% Source: SMMT Sterling depreciation raises new car prices In the aftermath of Brexit and sterling s rapid depreciation vs world currencies, numerous car manufacturers passed through swift price rises. Note that 80% of cars sold in the UK are imported. Price rises were passed through by; Ford (+1.5%), Honda (average +0.9%), AutoTrader (AUTOA.L) 6

7 Suzuki (+2% blamed on sterling) and PSA Group which owns Peugeot and Citroen (+2.5%). General Motors Europe stated that sterling s depreciation has already in effect cost the group 346m with its Opel CMO Tina Muller stating price rises will be the trend across the industry (Source: Autocar, 8 November 2016). The same AutoCar article reported that Hyundai would not be raising prices in a move labelled strategic by its UK CEO Tony Whitehorn. Despite not raising prices this year (2016) he stated "that s not saying that we will not raise prices in the future. There's a good discussion of price rises in this Autocar article. In addition to the price rises discussed (and, in some cases, implemented) in 2016, we would flag comments in early January 2017 from the CEO of the Society of Motor Manufacturers and Traders (SMMT), the industry's leading trade body, stating that UK consumers would see new car price rises of 2-3% over coming months. Rises in new car prices will lead to a reduction in the relative affordability of new cars for consumers which, in turn, will likely reduce new car sales, in our view. In the first instance, that may lead to used cars becoming better value in relative terms for consumers and some movement towards used car transactions from new car transactions. We would flag, however, that the availability of cheap credit means price rises would in many cases not be felt in a single lump sum, but spread over time. added to concerns over leasing cycle In our report entitled AutoTrader - Best in class but risks from cycle/automation (7 July 2016) we discussed the rise of financing in UK private auto sales (up to 86% in the 11 months to November 2016, from 46% in 2009) driven, in part, by captive financing (i.e. financed by companies' wholly-owned by OEMs) with which 37% of sales were financed in Figure 6: New car financing penetration rate for private customers Figure 7: Captive financing was used in 37% of UK car purchases (2015) 100% 80% 60% 40% 20% 86.6% 81.3% 76.1% 74.2% 71.2% 62.9% 45.8% 52.1% 47.4% ,200,000 1,000, , , , ,000 0 Non-captive 52% Not financed 11% Captive 37% News cars bought in finance (RHS) Private new car financing penetration (LHS) Source: FLA Source: AutoValue, Roland Berger, Credit Suisse research Our view is that captive financing has been artificially holding car prices up, as captive lending is struck at preferential rates, with an estimated 2ppt differential between captive and non-captive financing worth 3.2% on car prices. Those that finance their vehicles then, on average, own them for less time (c.4 years vs cash buyers at 6 years). We believe that this funding effectively artificially held up new car prices and led to the buildup of a significant stock of nearly new leased cars to come into the second hand market. The major build up started in 2011 and has continued through to As these cars come through into the used car market they'll likely pressure used car pricing and raise the relative affordability of used cars vs new cars. This will likely pressure new car sales. AutoTrader (AUTOA.L) 7

8 and further macro uncertainty The Credit Suisse Economics team forecasts UK growth to slow from 2.0% y/y in 2016 to 1.3% y/y in This is fuelled by slowing consumer spending growth, expected to fall from 2.8% in 2016 to 0.7% in 2017, in response to falling real disposable income, as the weaker sterling passes through and raises inflation. The team forecasts real disposable income growth to fall from 1.6% in 2016 to 0.1% in The magnitude of the impact on consumer spending depends on the degree to which consumers respond to the slowdown in income growth by borrowing more versus reducing their consumption this remains a significant uncertainty. The team believes the slowdown in consumer demand will likely be followed by a slowdown in the corporate sector. New cars are clearly typically more expensive than used cars so any pressure on consumer spending would likely reduce consumers propensity to buy a new car when other options exist (retaining existing car or buying used). leads to expectations for falling new car sales We believe the above three arguments support the seemingly wide held thesis that UK new car sales volumes have peaked. We flag forecasts from two leading external industry participants, with SMMT forecasting new car registrations down 5.0% in 2017, whilst Glass's forecasts the new car market falling by 3.5%. Comments from the listed franchise automotive retailers about market conditions after the EU referendum have been mixed. The most recent news flow has been particularly negative on the outlook for new cars. Marshall Motor Holdings stated in its FY16 pre-close statement on 12 January that it remains "cautious on the UK vehicle market in 2017 and concurs with current industry forecasts for a decline in the UK market for new vehicle sales". Along similar lines, Cambria Automobiles said in its 1Q17 trading update on 4 January "we believe that there may be some pressure on new car volumes and margins in 2017 as a result of the uncertainty in the economy and the foreign exchange volatility witnessed over the past few months." Vertu Motors has also highlighted the potential impact of the weakening sterling, saying at its interim results in August 2016 that "the market is starting to see vehicle price increases reflecting the manufacturers reaction to declining sterling exchange rates against all major currencies. Lower margin channels for manufacturers and retailers alike, such as Fleet car supply and Motability, are likely to see more impact than higher margin retail channels. There are diverging economic forecasts with regards to sterling s currency outlook and this leads to uncertainty over future manufacturer volume strategies and pricing." Contrastingly, Pendragon, the UK's largest retailer, seems less concerned about rising prices. Its Interim Management statement in October 2016 suggested limited pressure on pricing; "based on discussions with our franchise partners, we do not anticipate any material effect on new vehicle pricing as a result of exchange rates, since we believe the car manufacturers will try to mitigate this cost to consumers via monthly payments". We note that Motorpoint, the only listed independent used car retailer, stated at its interim results on 29 November 2016 that "despite the softening in consumer confidence, market conditions since the period end have remained stable with a good level of stock availability". This supports our thesis that, for the short to mid-term, used car sales should remain supported by the availability of off-lease stock. Used car sales lag new car sales, so will likely soften As shown in Figure 8, used car transaction trends tend to lag new car sales by about two years. This has been especially clear since 2010; since then, the correlation between annual new car sales growth and two-year lagged used car sales growth has been 82%. AutoTrader (AUTOA.L) 8

9 Figure 8: Annual new car sales vs 2yr lagged used car sales Figure 9: Monthly new vs used car sales, yoy 15.0% 15.0% 10.0% 10.0% 5.0% 0.0% -5.0% E 5.0% - Jan-15 Mar-15 May-15 Jul-15 Sep-15 Nov-15 Jan-16 Mar-16 May-16 Jul-16 Sep-16 Nov-16 Jan % -5.0% Brexit vote -15.0% YoY used car sales - 2 yr lag YoY new car sales -10.0% YoY new car sales YoY used car sales Source: SMMT Source: SMMT Before 2010, the steep decline in new car sales as a result of the 2008/09 recession was reflected more quickly by used cars sales, likely as a result of the significant contraction in consumer spending. Based on this historic two-years lagged correlation, we expect that growth in used cars sales will now begin to decelerate for the next two years likely turning negative around Used car prices matter too and they're slowing Given that used car retailers take ownership of their inventory (unlike real estate agents who solely act as agents) and aim to mark-up that inventory in order to make a profit, they are particularly susceptible to changes in end market pricing if they do not manage their inventory well. In such a scenario, retailers who do not turn their stock quickly enough or price conservatively enough could see profitability eroded. Figure 10: Growth in used car prices has been slowing since 2014 Figure 11: Monthly used car prices, yoy 12,000 10,000 8,000 7% 6% 5% 4% 10% 5% 0% 6,000 4,000 2, Average value per unit ( ) YoY change (RHS) 3% 2% 1% 0% -1% -5% -10% -15% Sep-14 Nov-14 Jan-15 Mar-15 May-15 Jul-15 Sep-15 Nov-15 Jan-16 Mar-16 May-16 Jul-16 Sep-16 Nov-16 Average used values (FLA) Auction prices (late & low, NAMA) Source: FLA, NAMA, Credit Suisse research Source: NAMA, FLA, Credit Suisse research. Note that the FLA data covers only vehicles bought on finance Note: NAMA data has been restated since Feb-16, therefore growth rates from Feb-16 are calculated from the restated base. Retailers always have a choice when selling their vehicles look to price more aggressively for a higher profit and risk a lower stock churn, or price more modestly and turn their stock more quickly. As shown in Figure 10, used car prices have been rising since 2013 and, bar two relatively flat years in , have risen since These rising prices have made the risks associated with miss-pricing relatively benign, with longer sale times really only causing lower stock churn and higher depreciation charges. AutoTrader (AUTOA.L) 9

10 However, this pricing vs time to sell decision becomes of crucial importance in an environment where used car prices begin to fall. Cars clearly depreciate on the retailer's forecourt but if the retailer is caught out by falling used car prices this can erode their profit margin and even force them into selling at a loss. Ultimately a prolonged period of falling prices coupled with poor forecourt management could lead to less sophisticated retailers going out of business. Autorola UK reported that the average price of used cars sold through its UK online remarketing portal fell by 2.5% in Q Figure 12:Average vehicle cost as a % of original value 70.0% 60.0% 63.2% 62.0% 59.5% 60.0% 61.9% 61.3% 60.2% 50.0% 40.0% 36.8% 37.0% 38.6% 39.8% 40.2% 39.6% 38.8% 30.0% 26.1% 26.2% 25.8% 27.2% 30.0% 29.7% 29.4% 20.0% 10.0% 0.0% 13.1% 13.5% 13.3% 13.7% 15.6% 16.6% 17.6% 3.4% 3.7% 3.7% 3.7% 4.2% 4.5% 4.8% YTD Nov years years years years >10.5 years Source: Glass's Rupert Pontin, Director of Valuations at Glass's discussed in his article "The end of an interesting year" published on 29 November 2016 (article here) that "volumes in the wholesale market have started to increase" and that this "is beginning to have a negative effect on used car values". Figure 12 above shows that both 'Late & Low' vehicles (0-2.5 years old) and 'Fleet profile' vehicles ( years old) have "begun to see a downturn". Pontin goes on to say that this is partly "attributed to the number of cars being defleeted". We see this as supportive of our view that the leasing cycle is and will continue to put downward pressure on used car pricing. Also as these vehicles flow through the car parc as older vehicles, it may also put downward pressure on the older categories (however, we note that older vehicles are surviving longer thus helping them retain their value). We would flag that the data shown in Figure 12 is particularly interesting for our thesis as it shows data closer to like-for-like used car pricing. Clearly if more higher value nearly new used cars flow into the market, used car pricing could be seen to rise, even if lfl used car pricing (or residual pricing) is falling. To be balanced, though, we flag that Motortrader.com reported in December that 80% of dealers are forecasting the used car sales market could remain flat or increase in 2017 compared to The article also cites one dealer who suggests that residual values of used cars could benefit from the price rises of new cars (article here). UK franchise dealers profitability is already being squeezed Our analysis below shows that UK franchise dealer profitability is already being squeezed (data from ASE) and that top independents (top 50 independents, data sourced from Motortrader) have been broadly flat over the last couple of years. AutoTrader (AUTOA.L) 10

11 In 2016, there was clearly a shift in blend from new car sales to used car sales for UK franchise dealers. This trend is shown in Figure 13 (for franchise dealers) and can be seen in Figure 9 for the market as a whole. We believe that used car gross margin ranges from c8.5%-25% with newer cars exhibiting lower margins and older cars higher margins. We believe that larger independent dealers likely average of around 9.7% in 2016 (Source: Motortrader Top 50 Independent Dealers report 2016). However we believe that smaller independents only operating in the used car market (no new and limited or nil aftersales/financing) typically selling older cars would operating on gross margins of around 15%. These figures are ahead the c.7% average gross margin obtained by major franchise dealers on new cars. Figure 13: A greater proportion of sales are used rather than new cars Figure 14: Absolute gross profit per new vs used car for dealers Gross margin per unit ,443 1,343 1, Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15 Jan-16 May-16 Sep-16 Ratio of used to new car sales (12m rolling) New Used (franchise) Used (independent) Source: ASE and Credit Suisse research Source: Company data (Cambria Automobiles, Vertu Motors, Lookers, Marshall Motor Holdings), Motortrader Top 50 Independent Dealers Report 2016, Credit Suisse research. However, the absolute profit per vehicle is lower for used vs new cars because the absolute cost of a new car is materially higher. For franchise dealers, we believe the average gross margin per unit is c. 1,440 for new vehicles vs c. 1,340 for used, based on data from Cambria Automobiles, Vertu Motors, Lookers and Marshall Motor Holdings. We estimate that the average profit of an independent used car retailer is c. 1,100 per vehicle sold (based on data from Motortrader Top 50 Independent Dealers report 2016). As such, we believe that a shift from new car sales to used car will cause a contraction in absolute profits of retailers as a whole. Franchise dealers already squeezed Figure 15 shows that whether as a result of the new to used shift, slowing new car sales growth or other factors, franchise dealer's razor thin net profit margins have already come under significant pressure over the last 2.5 years. Motortrader (see here) states average franchise dealer profitability was down 11% in 2016 in part due to a rise in self registration coupled with a reduction in underlying consumer demand leading to retailers squeezing margin in order to shift inventory. AutoTrader (AUTOA.L) 11

12 Figure 15: Franchise dealer net profit margin has been falling 1.60% 1.50% 1.40% 1.30% 1.20% 1.10% 1.00% Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15 Jan-16 May-16 Sep-16 Net profit as % of sales (12m rolling) Source: ASE and Credit Suisse research Independents solid at present Unsurprisingly given used car transactions have been rising and pricing is yet to turn negative, leading UK independent retailers are yet to see margins contract. In 2014 margins expanded from suppressed 2012 levels and have remained broadly flat since then. Figure 16: UK leading independents margins are broadly flat 3.20% 3.10% 3.00% 2.90% 2.80% 2.70% 2.60% 2.50% Source: Motortrader top 50 independent dealers report US trends support our thesis The US is ahead of the UK in seeing a transition in demand from new to used cars, largely fuelled by the availability of attractive off-lease vehicles. The US has also seen used car prices come off meaningfully. According to Manheim Consulting, off-lease units entering the used car market have been on an upward trend since 2013, increasing by 23% in 2016, from 2.55m to 3.13m. J.D. Power estimates that there will be a 6% increase in the total number of used vehicles 5 years old or less in the market in AutoTrader (AUTOA.L) 12

13 At its 2017 Convention and Exposition in January, the US National Automobile Dealer Association (NADA) warned of slowing new vehicle demand and softening used car values (Source: Wards Auto). Figure 20 shows weakening used vehicle pricing, -6% yoy to December 2016, according to the NADA Used Car Guide Used Vehicle Price Index. Figure 17: US auto market dynamics High off-lease volumes Shift of buyers from new to used Figure 18: US new light vehicle sales 20.0% 15.0% 10.0% 5.0% Price pressure on used vehicles Spiralling price pressure harms dealer profitability Pressure on new vehicle demand 0.0% -5.0% -10.0% -15.0% % Reduced effective new vehicle prices Auto manufacturers increase incentives for new vehicles -25.0% US new light vehicle sales yoy Source: Credit Suisse research Source: Autodata Figure 19: Off-lease units entering US wholesale supply Figure 20:NADA US Used Car Price Index Off-lease units (m) 114 Feb-13 Jun-13 Oct-13 Feb-14 Jun-14 Oct-14 Feb-15 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 NADA Used Car Guide Used Vehicle Price Index Source: Manheim Consulting Source: NADA So what does this all mean for AutoTrader? So, as discussed above, there are multiple different dynamics impacting the market for UK Auto retailers and therefore the market for AutoTrader. Our conclusion is that the ongoing shift in blend from new to used vehicles will continue to put some pressure on overall retailer profitability. However, with AutoTrader's disproportionate value-add coming in the used car space, we do not see this as a major issue. Given AutoTrader is partly priced dependent on stock, we believe rising used car sales will buoy AutoTrader stock, thus aiding ARPR. But overtime we expect used car prices to fall. Falling used car prices, we believe, will squeeze the profitability of some, less agile, retailers causing them to leave the market. Whilst stock will likely migrate to other retailers, AutoTrader will likely forego the fixed element of pricing (c.1/3 rd ). Overall ARPR will gain an additional tailwind (stock and blend) but this will not offset the incremental headwind felt by falling forecourt numbers. AutoTrader (AUTOA.L) 13

14 New car is more profitable than used car As discussed above in the section 'UK franchise dealers profitability is already being squeezed', we believe that the ongoing shift in blend from new car sales to used car sales has caused a contraction in retailer net profit margins with absolute profitability of a new car materially higher than that of a used car. We expect overall retailer absolute profitability to continue to be eroded by the shift from new car transactions to used car transactions. We would note, however, that given the greater proportion of new car sales concentrated amongst fewer franchise dealers and the lower prices they pay AutoTrader (we estimate they pay 0-40% less than group ARPA), pressure in this sub-sector may have a more limited impact on AutoTrader. We'd also flag that the nature of leasing a car could also limit the impact of rising vehicle prices on consumers as they would only feel a small proportion of the price rise on a monthly basis rather than in a lump sum. Rising used car sales buoy AutoTrader stock levels 97% of AutoTrader's stock on site are used cars with just 3% being new. We believe this is a reasonable indicator of the perceived relative value added by AutoTrader in the used vs new car markets. As such, even if falling new car sales erodes the absolute profit pools of UK auto retailers, their willingness to pay for AutoTrader services may well, at least in the first instance, be undiminished. With heightened stock flows into the used car market from the leasing cycle, we expect stock levels and transaction levels to continue to rise. With a proportion (undisclosed but we forecast 66%, see discussion below) of AutoTrader pricing directly derived from stock levels, rising stock will likely, in the first instance, lead to rising ARPR for AutoTrader. As such, we believe that FY17 ARPA could well be aided by some further stock tailwind. This would mean that both FY16 and FY17 were aided materially by rising stock levels. But eventually lower retailer profitability and falling used car prices will compress retailer numbers The most significant dynamic, in our view, is the impact that falling used car prices would have on the profitability of used car dealers. We believe that pricing will likely fall across Should this be the case, we believe that retailers' mark-ups will be eroded and those retailers not agile or adept enough (likely smaller independents) will be driven out of business. This will likely squeeze AutoTrader AutoTrader pricing is complex and not easily understood from an external standpoint. We know that the group prices across four product verticals; selling, buying, marketing and managing with the group then providing tiered products and services in each vertical. Some of these products are priced on a per forecourt basis and some priced dependent on the stock level. AutoTrader (AUTOA.L) 14

15 Figure 21: Pricing rises as retailers take higher levels and additional services RETAILER NEED SELLING BUYING MARKETING MANAGING LEVEL LEVEL LEVEL LEVEL LEVEL Stock Listing 880 Minimum Spend 1,240 Price driven entirely or almost entirely per forecourt Price driven by forecourt stock or advertised stock Minimum spend 1,240- price example, paid by a 25 vehicle customer for a four week Period Source: AutoTrader; however, note this is just illustrative and figures should not be seen as representative As such, if our thesis is correct and smaller retailers who are also less adept at forecourt management were to fold, the hit to AutoTrader would likely not be the entire revenue of those retailers lost. The folded retailer's stock would likely be hoovered up by larger or more agile retailers, and AutoTrader would only lose the fixed per forecourt pricing element of the folded retailers. AutoTrader derives c.80% of its revenues from Trade. Within that, around c.10% comes from Retailer Solutions (i.e. buying, marketing and managing verticals). Within Retailer Solutions, we believe c.50% of revenues are derived on a fixed per forecourt basis, with the remaining 50% on a per vehicle basis. Around 90% of retailer revenue comes from the group's Selling vertical. Within the Selling vertical, we estimate that around 30% of pricing is on a fixed per forecourt basis and 70% is dependent on stock. This revenue structure is outlined in Figure 22. These estimates are across AutoTrader as a whole; smaller retailers (those we deem to be most at risk) would likely have higher fixed component as a proportion of their ARPR. Figure 22: We believe c.1/3 of AUTOA's Trade pricing is effectively on a per forecourt basis with c2/3 on stock basis Retailer Solutions (by stock) 6% Retailer Solutions (per forecourt) 7% Selling solutions (by stock) 62% Forecourt based revenue 32% Stock based revenue 68% Selling solutions (per forecourt) 27% Source: Company data, Credit Suisse estimates AutoTrader (AUTOA.L) 15

16 To give some idea of AutoTrader's sensitivity; 1. Assume lost retailers have an ARPR of 1,536 p/m which is in-line with CSe FY17 AUTOA average. Whilst lost retailers will be smaller (lower stock component) they also likely have higher price points so around group average seems a reasonable working assumption. 2. Assume the above fixed/variable components (i.e 1/3 rd fixed and 2/3 rds variable), every 1,000 lost retailers on a FY basis would cost AUTOA 6m in revenues. This is around 2% CSe FY17 revenues. 3. Operational leverage would be high, assuming a 100% flow through, this 6m would be 3% CSe FY17 u/l op profit. Whilst this impact may not be particularly large, leading online classifieds players enjoy high multiples in part due to their visibility, consistent returns and upgrade cycles. Even relatively minor downgrades could cause meaningful de-ratings, in our view. According to our thesis, falling forecourt numbers at AutoTrader should be offset partially by stronger ARPR rises. We discuss our expectations in more detail later in the report (p20, Macro concerns could accelerate issues). Where could we be wrong? The two key areas we would flag are; 1) Used car pricing softens but doesn't decline Whilst we discuss multiple variables in the auto retailing market above, the key variable we see adding pressure to the profitability of retailers is used car pricing turning negative. Whilst we believe increased supply of used cars and softer consumer spending could put downward pressure on used car pricing, we could argue that rising new car prices and falling consumer spending would drive up used car demand and therefore support pricing. We acknowledge this as a risk. 2) Retailers find other efficiencies The second argument is that, even if used car pricing turns negative, retailers will be able to find efficiencies elsewhere in their cost base. Figure 23 shows that sales per salesman for franchise dealers has fallen materially over the last year implying that dealerships could well cut staff costs should they come under pressure. Whilst we acknowledge that larger retailers likely could make cost savings from their staff lines (and elsewhere), we would flag that remuneration for salesman is largely variable (BDO's Motor Retail Salary Survey 2015 stated that 60% of pay was from bonuses) so cuts would not necessarily save the retailer. We discuss this point in more detail in the section entitled "Retailer solutions need to deliver in FY18" below. AutoTrader (AUTOA.L) 16

17 Figure 23: Car sales per salesman for dealers has been falling Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15 Jan-16 May-16 Sep-16 Sales per salesman (12m rolling) Source: ASE, Credit Suisse research The key issue to monitor, in our view, to see if this thesis is playing out is the forecourt number at AutoTrader. Company collected consensus (as at 24 January 2017) is for AutoTrader to end FY17 with 13,402 forecourts, broadly in line with the 13,374 reported at HY and broadly in line with company guidance for forecourt numbers to remain flat through H2 (as stated at HY17 results). We expect to see 13,244 forecourts by y/e FY17. We would also note that, in June 2016, FY17 company collected consensus was for 13,607 forecourts. So consensus expectations for FY17 forecourt numbers have already come down materially. Whilst we are not below consensus for FY17 at a group level, we believe that any softness vs consensus in retailer forecourt volumes will support our view. AutoTrader (AUTOA.L) 17

18 Retailer solutions need to deliver in FY18 We have noted our support of management's strategy in previous reports (see AutoTrader - Best in class but risks from cycle/automation (7 July 2016)) particularly in relation to their innovative pricing strategy centred around up-selling incremental selling services and cross-selling other value-added services to retailers such as marketing, buying and managing services. In this section of the report, we look at the make-up of AutoTrader's ARPR growth and analyse how sustainable such increases may be. We conclude that the vast majority and a growing proportion of AutoTrader's ARPR rises come from underlying price rises and increasing stock levels at dealers. Whilst we acknowledge that AutoTrader services have made retailers more efficient, we believe that price rises centred on underlying price will ultimately prove to be finite and believe the group has a lot to prove with regards the potential for its Retailing Solutions products to deliver sustainable ARPR rises in future. Lastly, we also provide a scenario analysis. Using our proprietary analysis, based on calls with individual independent retailers and Automotive Consultant ASE's Chairman Mike Jones, we piece together what we believe represents average financials for a smaller UK independent used car retailer. We then analyse the impact of a reduction in gross profit margin (as a result of our cyclical concerns above) along with rising AutoTrader cost and find that under this scenario these dealers would likely fold. Whilst we do not believe this analysis is representative of all independent retailers, we do believe that should we see falling used car prices and further AutoTrader underlying price rises, some smaller independents who are less adept at managing their forecourts could fold (see Figure 30). Figure 24: Absolute YoY increase in ARPR ( ) by lever H H Price Stock Upsell Cross Sell Source: AutoTrader, Credit Suisse research Cross-sell a dwindling proportion of ARPR rises Whilst YoY percentage growth in ARPR accelerated in 2016 (over 2015) and we forecast it to remain strong in 2017 (over 2016), absolute rises in ARPR from up-sell/cross-sell haven't materially increased. The lion's share of ARPR rises have come from u/l price and stock. AutoTrader (AUTOA.L) 18

19 Figure 25: Proportion of YoY absolute ARPR rise by lever 42% 52% 39% 33% 15% 18% 34% 25% 39% 44% 27% 31% 19% 33% 48% H H Price Stock Upsell/Cross Sell Source: AutoTrader, Credit Suisse research At the group's FY16 results released on 9 June 2016, management guided that it expected high single digit growth in ARPR for FY17, with this growth split broadly equally between price, stock and up-sell/cross-sell. H ARPR growth was very strong at +13%; however, the beat was driven almost entirely by higher underlying price rises. Whilst this strength in underlying pricing clearly underscores the group's strong competitive position and pricing power, we believe that pricing power is ultimately finite. Auto retailers only have a finite profitability that AutoTrader can target. Growth in cross-sell is indicative of AutoTrader adding wider value to retailers which we deem to be more sustainable. Whilst classified advertising partners such as Rightmove, AutoTrader and ZPG always provoke divisive views from end customers, we would flag that our analysis of the independent dealers highlighted a growing dissatisfaction around AutoTrader's buying vertical, Autotrad (a platform to help retailers source cars from within the industry). Underlying price rises can't go on forever We estimate that, in 2017, online auto classified advertising will account for c.86% of total auto classified advertising. This is relatively high vs other classified categories, for example property stands at c.79%. Figure 26: Credit Suisse estimates of UK Auto classified advertising ( millions) Figure 27: Credit Suisse estimates of UK Property classified advertising ( millions) National Newspapers Regional Newspapers Total Magazine Total Online National Newspaper (print) Regional Newspaper (print) Rightmove ZPG Other Digital Source: WARC, AutoTrader, Credit Suisse estimates and research Source: WARC, Rightmove, ZPG, Credit Suisse estimates and research AutoTrader (AUTOA.L) 19

20 If the vast majority of auto classified advertising goes online, as forecast in Figure 26, in order to pay AutoTrader more on underlying price, retailers will either need to find cost savings elsewhere in their business or they'll need to erode their profit margins. Growth in cross-selling products is typically, in our view, accompanied by cost savings elsewhere in retailers' businesses and as such is sustainable. Price hikes are more likely to simply erode retailer profitability. Macro concerns could accelerate issues We would argue that online market places such as AutoTrader have added value to their customers beyond the provision of classified advertising services, allowing retailers to improve their efficiency around technology, sourcing, pricing and headcount. So one could argue that retailers potentially 'should' spend a greater proportion of gross profit on classified advertising now than they have historically. Whilst we do in part believe this theory, we believe that, for some retailers, macro concerns could squeeze retailer profitability forcing some retailers out of business. Finding granular data on the Auto retailing market is difficult, and finding data on cost breakdown and margins is particularly hard. We look in detail at the margin profile and cost breakdown of leading UK franchise dealers and spoke to a number of industry participants and specialists in order to estimate the margin profile and cost breakdown of average independent retailers across the UK. These retailers are the backbone of AutoTrader's business. In pulling together the following analysis, we spoke in depth with the owners of five independent dealerships and with ASE Chairman Mike Jones, in order to build up an informed view of the average cost structure and margin profile of independent dealers. ASE is a leading automotive consultancy specifically focused on dealer profitability, transactional management and provides commentary on key industry issues. Figure 29 shows our estimates for average gross profit margin and breakdown of gross profit between different cost buckets and operating profit, for the UK's independent used car retailers. In this scenario, i.e. the current operating environment, there are clearly cost buckets and profitability AutoTrader could target through price rises. Figure 28: Assumptions for 2019 Scenario vs current 2016 scenario 2016 base case scenario Assumptions for 2019 potential scenario vs 2016 base case 2019 potential scenario* Gross margin (% of revenue) 15% Gross margin falls to 11% 11% Costs as % gross profit AutoTrader 12.0% +10% p.a annual price rise 21.8% Other classified advertising 3.0% -25% p.a decline in other classified spend 1.7% Overheads (incl property) 28.2% -3% p.a reduction in overhead costs 35.0% Staff 33.7% -4% p.a reduction in staff costs 40.6% Operating profit 23.2% 0.9% Operating margin (% of revenue) 3.5% 0.1% Source: ASE, SMMT, AutoTrader, Credit Suisse estimates and research *Note: "2019 potential scenario" is a scenario we believe could be a reality for some, less agile retailers. We do not expect this scenario to playout across the entire dealer market. AutoTrader (AUTOA.L) 20