Description

AutoNation is the largest auto dealer in the US, operating 349 franchises at 252 locations (they also have some collision centers and a handful of used car only locations). We have been shareholders of AN for most of the past twenty years. When we first bought AN shares, we thought the parts and service (P&S) business was misunderstood, the overall business was less cyclical than other investors believed, that finance & insurance (F&I) had room to grow, the impact of floorplan financing made the businesses look over-leveraged, and that good capital allocation would provide a boost to our underlying economic returns. Before I get into AN specifics, I would point out that we think several of the auto dealers are interesting now. ABG is an example (cubbie, cupmachine314, and manatee did terrific work fairly recently on ABG, so we thought VIC didn't need another ABG note).

How did our original AN thesis play out?

1. Today, everybody understands the solid economics of the P&S business, and the gentle mix shift toward overall higher EBIT margins that P&S provides to the entire business. Twenty years ago, P&S was 36% of total company gross profit, while today it is 41%.

2. The business has proven to be less cyclical than one might think given the performance of OEMs over the past twenty years. Indeed, AN posted a profit in 2008 (excl goodwill impairment) and 2009 (and at that time, a number of theses floated around Wall Street that AN was imminently bankrupt -- including by my boss at that time, fun times... but I digress).

3. In 2003, AN reported F&I revenue per car of $915. During management visits at that time, we quizzed the company on whether/how they might get F&I to crack $1,000 per vehicle. Keep in mind, F&I revenues are 100% gross margin, so growth in F&I contributes to the overall EBIT margin boost over time. In 2023, F&I per vehicle was $2,736 (growth of 5.6% per year for 20 years).

4. Again, going back to 2003, investors seemed confused about the actual debt levels at the auto dealers. OEMs (and now other floorplan lenders) provide financing for dealers to carry inventory. In most cases, floorplan financing is a rough offset to inventory (especially new car inventory). OEMs provide incentives to move metal, and those incentives are recorded in new car gross profit. The floorplan expense depends entirely on the level of interest rates, and how long a vehicle remains on the lot before it gets sold. In many years, AN recorded a net floorplan profit (the OEM incentives were higher than the interest expense on floorplan financing). However, the floorplan debt shows up on the balance sheet making leverage look high. The key here is that floorplan is tied to inventory... it's effectively the cost of working capital (and it's usually a profit), not extra leverage. Today, that is better understood than it was 20 years ago.

5. Capital allocation has been terrific. In 2003, the company had 269.7M shares outstanding. Today, there are 41.6M shares outstanding. Effectively, the company has repurchased 85% of its shares over the last 20 years, or 8.9% per year. They have never paid a dividend.

Put all that together, and if you bought shares 20 years ago and held them, you made ~12% annually without paying a dime of tax. There have been plenty of opportunities to add (and trim) if you were so inclined. I'm suggesting that now is not a bad time to add.

Where are we today?

It is not exactly easy to say what "normal" is for the auto dealers today. In the last 5 years, we've had COVID, enormous supply chain disruptions, rapidly rising interest rates, inflation, used car prices through the roof, and a consumer today that in surveys claims to be quite concerned about the economy. As a result, very little about the financial statements in any of the last 3-4 years can be thought of as normal. Put another way, what might look really good on one line of the financials likely has an offset somewhere else.

We would argue that the overall business is about the same quality as it was 20 years ago, but again, there are offsetting factors. Our view is that the P&S business is a bigger part of the overall business, and that P&S is an above-average part of the dealership business model. Also, F&I is bigger and better. The new car business is probably less interesting primarily, in our view, because of internet search. A car is a big purchase, so consumers spend time trying to get a good deal. It is not much effort to lob in a request for best price to several nearby dealers and force them to compete for your business (and since a new car sale produces F&I income and perhaps a P&S annuity, you can sacrifice margin on new to drive those other lines of business). We would also argue that the rise of Carvana, etc. has made the used car business less attractive. Again, this basically comes down to ease of gathering information and thus being a more informed consumer. The good news in used is that there are tax advantages to trading in your older vehicle when you buy a new one rather than selling it to KMX or CVNA. Still, it is hard to argue that the used car business is as good as it was 20 years ago.

The bottom line is that returns on tangible capital are still very solid for auto dealers, especially those that operate at scale.

What is is worth?

In 2023, AN sold 244,546 new cars and 274,019 used cars (518,565 total units). Total US new car sales were 15.5M. US car sales at that level seem reasonable, though the average age of the US light vehicle fleet gets older at the 15.5M SAAR. We think new and used car margins are probably permanently lower than they were historically. If we look out over the next five years, we think the EBIT line might look something like this:

New car gross profit  = $1.2 billion

Used car gross profit = $500 million

P&S gross profit = $2.3 billion

F&I gross profit = $1.5 billion

Total gross profit = $5.5 billion (in 2023, it was $5.13B)

SG&A = 62% of GP, or $3.4 billion

D&A = $250 million (and is roughly equal to maintenance capex over time)

EBIT would be ~$1.8 billion

Interest ~$200 million (we include floorplan interest expense in the new/used vehicle margins - current debt has low rates and maturities are several years out)

EBT = $1.6 billion taxed at 25% 

Net income = $1.2 billion

They will buy back at least 10% of shares annually. In five years, there is a good chance that AN has less than 25M shares outstanding so EPS would be around $50/sh. We think the business quality is about the same as today and deserves at least a 10x multiple (the stock consistently traded above 10x for many years, but perhaps a lower multiple is the new reality. We'll stick with 10x which gets a stock price of $500 in five years.

Risks:

1. AN has recently started their own lending business where they keep some of the loans they originate (the "non-recourse" debt on their balance sheet is the financing mechanism for this business. It is a small part of the business today, but adds risk to the business as it grows (at least in our opinion). I'd prefer that AN not do this, but for now we'll live with it.

2. My personal opinion is that current management isn't as good as prior management teams. This may not prove to be true, but that is my current take. There are Board members that seem like they understand shareholder value, etc (like Bob Grusky, etc), and the capital allocation plan remains intact, so perhaps my concerns are unfounded. 

3. Quick transition to EV -- we think this is very unlikely, but it would impair the P&S business if it happened. For now, the P&S business is benefitting from a rising average age of cars in the US. Older cars need more service.

4. For shorter-term owners, keep in mind that the Gates Foundation and Eddie Lampert combined own ~34% of the outstanding shares. Should they choose to sell, that might be a near-term problem for the stock.

Catalyst

No specific catalyst other than time and patience.