Lennar Share Class Arbitrage $LEN

In my annual letter, I included a section on finding opportunities to take advantage of the rise of indexation. I know that simply saying “we are taking advantage of increased indexation” is a bit of a nebulous statement and something that almost every small / value fund manager says; in general, they simply mean “we invest in companies that aren’t in the indices.” That’s certainly true to a large extent for me too, but I think there are a few investments we’ve made that are designed specifically to take advantage of the rise of indexation that separates us from such broad stock talk. To show that, I wanted to highlight a specific example to show how we look to take advantage of indexation, so let me present my favorite “take advantage of the index” idea today: Lennar share class arbitrage (disclosure: Long Lennar B shares, short Lennar A shares).

Some background: Lennar is one of the nation’s biggest home builders and traces its roots back to the 50s, and the current CEO, Stuart Miller, is the son of one of the co-founders. What interests us here is that the company has a dual share class structure where class A shares get one vote per share and super voting class B shares get 10 votes per share (the difference in votes is the only difference between the two share classes; they are identical in all other aspects; see p. 17). This dual class structure has allowed the Miller family to retain effective control of the company: until recently, the Miller family owned ~10% of the company but had ~42% of the vote (see 2016 10-k p. 17) thanks to their ~69% ownership of the super voting B shares (see 2017 proxy p. 47).

In general, a super voting class of shares should trade for a premium to the single voting shares, not a discount. As a specific example, Heico (HEI) has two classes of stock similar to Lennar (one class gets 1/10th of vote per share while the other gets a full vote per share). Heico’s full voting shares currently trade for a 20-25% premium to the 1/10th vote shares (~$90/share versus ~$71/share). Yet, as I write this, Lennar’s super voting class B shares trade for ~$48/share while the single vote class A shares trade for ~$60/share, or about a 20% discount. I believe the discount is driven by liquidity (several million A shares trade hands daily while only ~50k B shares trade each day) and index inclusion (the A shares are included in the S&P 500 index while the B shares are not). This discrepancy presents us the opportunity to go long the B shares at a discount while shorting the A shares. Our net economic exposure here is effectively zero, and if the gap ever collapses we will make a >20% arbitrage profit.

Of course, “if the gap ever collapses” are some of the most dangerous words in finance. However, we believe that we could be on the verge of an inflection point where the gap could quickly collapse. Lennar recently issued a dividend of class B shares to both class A and class B shareholders, and they recently completed a merger that had them issue both class A and class B shares to the acquired firm. With the merger complete, the Miller family vote is down to ~34% of the total vote (see p. 55).  As the family’s share of the total voting block goes down, the odds that an activist steps in and successfully proposes a collapse of the voting structure increases. Activist Gamco has been  doing exactly that over the past few years, and this year they are again proposing giving the ability for Class B shareholders to convert to Class A (see proposal  6). If the huge spread continues to persists, we could see increased noise from other activist / event funds looking to play the spread and force a collapse (i.e. a larger version of what we’re doing). It makes sense too- it is in everyone’s (except for the Miller family) best interest to collapse the spread. Class A stock becomes more valuable when the super voting shares are eliminated, and all outside class B shareholders benefit from a more liquid stock that sells for a significant premium.

There are a lot of other things we like about playing the Lennar spread. Because the class A shares are in the S&P 500 and extremely liquid, shorting the A shares is easy. And we love owning the super voting shares because we own the exact same thing that management owns, so there’s a chance the B shares end up getting a premium in a merger or that A shareholders vote to pay B shares a slight premium to remove the super voting overhang (we’ve seen this done a few times; Stewart in early 2016 serves as one example). There’s also an extremely unlikely scenario where an activist runs an activist campaign on Lennar by buying up a huge chunk of B shares, in which case it’s possible we could exit the B shares at a healthy premium to A shares.

Are there risks? Sure. In the short term, we could have our short taken away from us (extremely unlikely given the company is in the S&P 500, but certainly possible in a 2008 like scenario), or the spread could blow out on us and subject us to mark to market losses. But, over the long term, we are buying a superior economic asset in a perfectly hedged manner, so it’s difficult for us to see a scenario where we have a permanent loss of capital.