|Shares Out. (in M):||187||P/E||17.67||19|
|Market Cap (in $M):||12,624||P/FCF||0||0|
|Net Debt (in $M):||1,342||EBIT||1,006||1,006|
|Borrow Cost:||General Collateral|
A lot is baked-in here after decent numbers. CarMax has been a disruptive force in the used car industry with its “no-haggle” pricing, transparency, and customer friendly business model. Since inception, the company been able to grow and win share in an established, fragmented industry. However, this well understood by consensus and is priced into the shares. A differentiated perspective illustrates the fact: 1) competitors will soon eat into profits, 2) the rise in interest rates won’t help; 3) the company lacks operating leverage. These issues are overlooked by the marketplace:
1) Competition set to increase from existing competitors and new entrants. Perhaps the biggest secular risk facing CarMax is the potential of new and emerging competitors, which appear to be copying the company and finding innovative ways of doing business.
• AutoNation USA – Launching a similar concept to CarMax that will sell used vehicles with no-haggle pricing, although in slightly smaller formats. The company announced their plans as part of AutoNation’s new brand strategy two months ago and is planning on opening 5 stores in 2017 with another 20 stores identified in existing markets
• EchoPark (owned by Sonic Automotive) – Another retail concept similar to CarMax with no-haggle pricing and great customer service. The company currently has 5 stores in Denver (one more set to open soon) with future expansion set for Texas and the Carolinas. Sales this year are expected to be $120M with 5k units sold
• Q Auto (owned by Asbury Automotive) – Currently has 4 stores in Tampa in a small format model and recently closed a large store. Company won’t open more stores until it figures out how to increase profitability
• Carvana – An internet based, startup used car retailer primarily in southeast regional markets, but looking to expand aggressively (11 markets opened this year in addition to 9 existing markets). Company has raised $300M in equity and $400M in debt and is expected to generate $350M sales this year. Customers can buy a car online and get it delivered for free (within 100 miles), or alternatively, pick it up from an eight story vending machine. All cars are Carvana certified and come with a 7-day no questions asked, money back guarantee
• Vroom – Similar to Carvana, an internet based used vehicle retailer with free delivery (anywhere) and 7-day test drive. Company has raised $250M (the most recent round was led by T Rowe Price) and is expected to have $1.1B in revenue this year selling 50K vehicles, increasing 22% from last year
Worth noting is that all of these competitors generally target late model vehicles, CarMax’s focus area, and the lack of a physical footprint for many of the online startups could provide them with the ability to offer lower prices. Even though the combined sales are all fairly small relative to CarMax, if growth rates are maintained, they could be a formidable competitor years later.
2) Higher interest rates to pressure CAF income – Since the election, the 3 year treasury note (which is a proxy for interest rates on securitizations given the average length of a loan is over 2 years) has increased nearly 50bps from 1.0% to 1.5% and nearly 70bps from the last securitization in July. CarMax could choose to pass the increase in costs directly to consumers, but that higher cost would likely serve to hamper demand by making vehicles effectively more expensive. Within the last rising rate environment witnessed in 2003 to 2006, CarMax passed on roughly 35% of the increase to consumers in the form of higher APRs (Exhibit 1). With the prospect of continued rate hikes next year and in the future, CAF net interest margins will likely continue to decline moderately, leaving segment income relatively flat over the next few years (as increased loans originated and a higher average A/R balance help offset the decline in margin).
3) Lack of operating leverage – One of the biggest drivers of earnings growth (and the stock price) is change in same store sales. Since 2003, the correlation coefficient (r-squared) between the two has been .80. Given the drag on profitability from new store openings which are burdened with pre-opening expenses and a ramp to maturity, the business requires comps in the 5%+ range to realize operating leverage. With decent, yet unimpressive comps forecasted for the next few years given where we are in the cycle and a lack of growth in the SAAR, I think CarMax will generate 7% annualized EPS growth through FY 2020 (2% net income growth excluding the buyback impact on EPS), hardly an impressive figure for a growth retailer.
In building a model, the following set of assumptions were made on key drivers to derive earnings over the next three years (through FY 2020), which ultimately lead to estimates below consensus figures.
• New store openings – 15 to 16 new stores annually, per management guidance and confirmed by a bottoms up analysis on store potential by MSA market
• Used vehicle sales – Historically, there has been a general relationship between new and used car sales that are often driven by the economic environment, as both are large big ticket items that generally require income, employment and confidence. Since 2008, the r-squared between CarMax’s used comps and the SAAR has been .50, not significantly high, but directional nonetheless (Exhibit 2). Estimates of a flattish SAAR would imply flat comps, but CarMax should outgrow this due to new stores ramping up (1/3 of their store base is comping 7.5% annually which implies a ~200bps comp tailwind), growth in off-lease vehicles returning to the market (500k incremental off lease vehicles should be returning to the market annually over the next few years, which makes used vehicles more attractive relative to new vehicles through lower prices), as well as a modest benefit from a better SUV/truck mix in the short term, resulting in 3-4% growth in comps. Gross profit/vehicle expected to remain flat, as has been the trend over the last 6 years (Exhibit 3), despite increased internet transparency. Management seems only willing to give up margin/unit if it drives additional unit sales and more total gross profit dollars
• Wholesale sales– Units decline low single digits next year due to lack of 7-9 year old vehicles offset by new store growth with margins under modest pressure from lower ASP’s (greater mix of even older vehicles which will weigh on ASP) and sales volumes.
• CAF – margins come under pressure from higher rates, only which a fraction is passed through to consumer in the form of higher APRs, with penetration expected to be flat at 45%
• SG&A – Salaries, commissions and benefits trend in line with growth in used vehicles sold, advertising grows in the low single digits, occupancy increases based on store growth, and overhead continues at elevated levels given the company’s continued strategic focus on website/mobile optimization, testing home delivery, and building online financing capabilities. Together, results in slight SG&A/vehicle deleverage, going from $1,398 per vehicle to $1,472 per vehicle.
Three different scenarios are modeled, with key assumptions modified and price targets derived off of P/E multiples, as the market seems to be using that as their valuation methodology. Base case assumes steady growth continues, much like FY 2017 results. Downside case assumes credit availability tightens impacting demand, gross profit per unit weakens due to weak macro environment and heightened competition, and loan loss provision increases to 2%. Upside case is predicated on a 25% tax rate from corporate tax reform and an acceleration of comps to 5%. Given the threat of new competition, I’d be more inclined to be short the stock, but would want for a higher price in the mid $70’s to ensure at least a 2:1 risk/reward profile.
A majority of customers (70%) purchasing a vehicle from CarMax will obtain in store financing. When a customer requests financing it is first routed to the company’s internal finance division, CarMax Auto Finance (CAF), which will present a variety of financing options to choose from (assuming the customer is credit worthy, average FICO stores within CAF are in the low 700’s). If the customer gets rejected from CAF the application is then routed to a Tier 2 credit partner (Wells Fargo, Capital One, Ally, etc.) and the customer can choose one of their offers, after which CarMax gets paid $250-$300 for generating the lead. If a customer application is not approved at this point (and hence is a high risk, subprime credit), it is next routed to a Tier 3 partner (Santander or Exeter) and if approved, CarMax pays the Tier 3 firm $1,000, which it is willing to do because it generates the $2.2K in gross profit from the sale of the vehicle. Worth noting is that CarMax is currently doing a test and underwriting a small portion (70bps of the overall receivables balance) of Tier 3 sales, in order to avoid having to pay the $1,000 fee and drive better profitability among this group. Nearly 45% of vehicles sold are financed through CAF, 15% through a Tier 2 partner, and 10% through Tier 3 (with 30% paying cash or bringing external bank financing). For each loan generated, CarMax makes approximately $400 annually over the life of the loan (including an allocation of store SG&A expenses, see model for details).
Typically every 3-4 months CAF will securitize a pool of receivables in the auto ABS market (with a warehouse facility being used in the interim). Essentially, the company will place the receivables into a trust and the trust will then sell notes to investors. Cash flows (principal and interest) from the loans go into the trust, which are then used to pay the investors their principal and coupon payments and the remaining interest payments (above the interest expense) goes back to CarMax. It is important to note that these notes are non-recourse to CarMax and risk of principal default lies with the investor (although CarMax would suffer from less interest income in the event of a customer defaulting). In the latest securitization (2016-4), the company’s average APR was 7.34% and average interest expense was 1.5%, leaving CarMax to capture a 5.84% spread (before loss provisions and expenses). Historically, in the last 41 securitizations since 2003, the average spread has been 6.0% with a median of 5.7%. Captive financing is a good business for CarMax as it allows the company to generate additional vehicle sales and drive cash flow in an asset light manner.
In 2015 approximately 39M used vehicles were sold, with sales roughly split evenly between 18k franchised dealers (37% share by volume), 36k independent dealers (33% share), and private party consumer-to-consumer transactions (30% share). The industry is heavily fragmented, with the top 50 franchised dealership groups only selling 1.7M used vehicles last year. Of the ~39M used cars sold, ~22M were within the 0-10 year old range, CarMax’s focus area, putting them more in competition with franchised dealers as independents typically focus on older vehicles. CarMax sold 620K used vehicles last year (FY 2016) and has a 3% national market share – however, given that CarMax isn’t in all locations, the company believes it has closer to a 5% share in the market it operates in and a 10% share in its most mature markets such as Los Angeles, Chicago, and Washington DC.
Supply of used vehicles comes primarily from customer purchases of new vehicles (leading to trade-ins) and vehicles coming off lease (and to a lesser extent from rental car companies and fleets). Given that CarMax focuses on late model vehicles and that ~80% of their inventory is in the 0 to 4 year old range, the best way to measure supply is through changes in the above variables. Similar to new vehicle sales (SAAR), demand is driven primarily from macroeconomic factors (GDP growth, employment levels, consumer confidence), the age of the installed base, credit availability, and affordability relative to new cars.
CarMax generates profits from three main functions – selling used vehicles wholesale, selling used vehicles retail, and providing financing for customer purchases. The process starts with appraisals where a customer can sell their car to CarMax. An associate will look at the vehicles’ history, perform an inspection, and conduct a test drive. Based on the condition (primarily age/mileage but also make/model) and real time data on what similar make/models are selling for, a seven day written offer will be provided to the customer, which is independent of the customer purchasing a car. If the offer is accepted (which happens 30% of the time), but CarMax doesn’t want to sell it retail (typically if it is over 100k miles or 10 years old) the company will conduct its own auction and sell it to independent dealers, who know the car has already been vetted through the CarMax appraisal process. A markup is made on the sale in addition to dealer fees that are collected, which are proportional to the amount the dealers spend. Last year, 390k cars were sold wholesale with an ASP of $5,500 and a GP/unit of $980. If the vehicle is a younger model one in good condition CarMax will sell the car itself through retail (in addition to vehicles it purchases from auction), but before doing so, will recondition it to bring it to its quality standards. Last year, nearly 620K cars were sold retail with an ASP of $20.1K and a GP/unit of $2,160. Along with the sale of the vehicle, customers will be offered extended service plan warranties and asset protection plans (administered by third parties), which generate high margin revenues (an average of $440 on every used vehicle sold). Lastly, an integral and lucrative part of the process is offering financing to customers through the company’s captive finance arm (described next).
Negative preannouncement a few weeks before;
Analyst Day April, 19th