We think AutoNation’s core economics are generally misunderstood by the investment community for a couple of reasons.  First, let’s take a look at the balance sheet.  As of December 31, total assets were $8.5 billion.  Of that, approximately $3.1 billion is goodwill/intangibles generated as Wayne Huizenga used very high priced shares a decade ago to build the company.  Another $2.3 billion is inventory.  This is what makes car dealerships unique.  Practically speaking, there is very little inventory risk with new cars.  The OEMs finance the product through a floorplan program (more on that later) and the retailers effectively have the right to “put” the inventory back to the OEMs.  Operationally, this rarely happens because if new car sales run into trouble, the OEMs generally introduce rebates or financing plans to move the inventory.  As a result, unlike most retailers, inventory risk is extremely low.  Effectively, the true capital employed by AutoNation is just over $3.2 billion of which a sizeable amount (currently around $750 million) is offset with non-interest bearing current liabilities.  The capital structure, and returns on capital, is far more interesting than it appears at first glance.

Now, let’s turn to the income statement.  As you might expect, the majority of revenues (~58%) come from new car sales.  Roughly 35% of AN’s new vehicle sales are domestic nameplates, 40% are imports (Honda, Toyota, etc.) and the remaining 25% are luxury imports (BMW, Mercedes, Lexus, etc).  Gross margins on new vehicle sales hover around 7%, though that does not include the salesperson commissions which are generally about ½ of the gross margin.  Automotive News recently reported that the average dealer of domestic cars in the US lost money on each new car sale (there is a fairly wide variance among dealers, but it is important to note that the average domestic dealer lost money on each new car sold last year).  One thing is clear, the United States has far too many domestic nameplate dealerships (more on this later).  Used car sales represent ~24% of revenues, and typically have gross margins of 8-10%.  AutoNation has an interesting advantage when it comes to used cars.  Their strategy is to essentially sell every brand in a given metro market.  So, if you trade in your Toyota Camry for that new Lexus, they are very likely to move your used Camry to their local Toyota dealership where it is much more likely that potential used Toyota buyers will shop (their scale also allows customers to search all of the local AN dealerships for a specific used car via the Internet). 

Parts and service (P&S) revenues are about 15% of total sales, and carry 43-44% gross margins.  This is the gem within AutoNation.  Roughly a third of revenues in P&S come from warranty work.  The OEM issues a recall, and typically pays each dealership a fixed amount per car serviced on recall work.  Ideally, a dealer can assign a lower paid tech (techs get paid big $’s) to these jobs and the tech can climb the learning curve quickly such that the hours allotted by the OEM are significantly outperformed on each job (and thus they earn very high margins).  Also, nowadays, most service shops put your car on a computer for diagnostics.  By law, the OEMs must sell the diagnostic software to all comers, but practically, it is much cheaper to run that fixed cost over 40 service bays at the local Mercedes dealership than to have a mom-and-pop outfit that services all sorts of vehicles.  The diagnostics software isn’t cheap.  Given some of the warranty and maintenance plans extended to buyers, along with service agreements sold with the new car, the customer has a huge incentive to get service at the dealership.  For example, BMW sells all new cars with 4-years of no maintenance cost.  All BMW owners would be foolish not to have their work done at the dealership.  The complexity of new cars is slowly building a barrier to mom-and-pop service shops and do-it-yourselfers.  P&S is truly a gem of a business for AutoNation.

Last, but not least, finance revenue makes up 3-4% of total sales.  AutoNation provides third party financing for customers, but does not accept any credit risk.  Approximately 2/3 of car buyers need financing, and this gives AN an opportunity to sell extended service agreements, and other items (anti-theft features, etc.).  Finance revenue is 100% gross margin, because AutoNation essentially gets a fee for selling the product or originating the loan for a financial institution.  Importantly, there is a significant gap between top quartile stores and bottom quartile stores when it comes to finance revenue per car.  Most future opportunity lies in closing some of the performance gap among stores.

As I mentioned earlier, the floorplan financing arrangement makes auto retailing unique.  Let me quickly walk through the mechanics of floorplan.  Upon receipt of a car from the OEM, the dealer has a liability for the cost of that car in the form of a floorplan note payable.  That note accrues interest just like any other kind of borrowing.  However, the OEMs extend “floorplan assistance” to dealers to offset this floorplan expense.  While each nameplate is different, and floorplan programs change to meet the desired sales results of the OEM, a typical plan might include paying the dealer for the first say 60 days of floorplan expense.  So, the dealer has the car on the lot and earns a “floorplan benefit” if he can sell it in less than 60 days, but has a net “floorplan expense” if it takes longer than 60 days.  Again, the OEMs tweak these plans to meet their sales goals.  If GM wants dealers to buy more trucks, they may extend the assistance out to say 120 days.  Some OEMs provide a flat fee per vehicle (if you sell a Toyota the day you receive it, your profit rises by that payment).  As a result, the Floorplan Notes Payable on the balance sheet is typically financed by the OEM over time.  Historically, it made sense just to borrow from the OEM, because you essentially could get their credit rating for your interest cost.  In some cases, it now may make sense to finance the floorplan through a third party and create a wider spread between floorplan benefit and floorplan expense.

This is a key point.  Over the past 5 years, AutoNation has generated nearly $24 million in pretax profit from floorplan benefits.  However, during the past four quarters, they have LOST $33 million pretax on this program.  In other words, before this past year, they were ahead on floorplan by roughly $57 million in the past 4 years.  With the inclusion of this past year, their 5-year cumulative floorplan benefit shrinks to a total of $24 million.  In other words, the floorplan program has cost AN about 15 cents per share pretax in the past year.

As a result of the floorplan program, we think it makes sense to think about AutoNation’s operating income excluding floorplan.  Reported EBIT for the last 4 quarters (through Dec 31) is $706 million.  Adding back the floorplan cost for this past year, the adjusted EBIT is $740 million.  Current valuation is as follows (assuming a $16.30 share price and the sharecount from the most recent 10K given their heavy share repurchases of late):

The latest K has shares outstanding (Feb 26, 2008) of 180.0 million shares.  There are 14.2 million options outstanding with a wtd avg strike of $16.70 (so not very dilutive at the current price).  As a result, the market cap of the company is $2.9 billion and net debt is $1.7 billion for an enterprise value of $4.7 billion.  Using the adjusted EBIT discussed above (excluding the impact of floorplan), EBIT yield is just a touch under 16%.

In April of 2006, AutoNation did a tender offer for about 25% of their stock at $23.  They introduced leverage on the balance sheet to accomplish the tender.  Management has stated that their internal return hurdles are 15% after-tax on capital allocation decisions.  Given the debt covenants, they can essentially use half of their net income each year to repurchase shares (it’s more complicated than that, but this is conservative).  Roughly half of capex is maintenance, and the rest is generally for expansion in P&S. 

Where will they spend the rest of their cash?  One thing is clear.  There are far too many dealers in the US selling the domestic nameplates.  GM, Ford, and Chrysler are all working toward shrinking the dealer footprint.  Given the history of state franchising laws for car dealers (written when the Big 3 dominated auto manufacturing), the OEMs cannot force closures.  A significant consolidation is in order.  In fact, estimates are that the Big 3 OEMs have at least 20% too many dealerships (we would argue that figure is too low).  Just to give this some flavor (using very rounded figures), the average Big 3 dealer sells 600 cars per year compared to say 1,500 for a Toyota dealership.  The domestic OEMs are trying to consolidate brands under a single roof, working with dealers to promote consolidation, and may wind up paying money to some dealers to close their doors.  As an aside, in quite a few instances, the land underneath the dealership in some markets may be worth more than the dealership’s earnings stream.  Most likely, this consolidation will be a very slow process, but AutoNation is in a clear position of acting as the consolidator especially if the current weakness in auto sales persists.  They are the low-cost provider with the best access to capital and the best operating system in the business.  We suspect that the company will put the remainder of its cash flows toward acquisitions over the next couple of years assuming they clear the 15% after-tax ROIC hurdle.