Marc Andreessen - currently near, and possibly at, the top of my “favorite bloggers” list - has a fascinating post today praising the virtues of dual-class stock structures. The really short version: they’re good because they allow the founders (or whoever has the voting power) to ignore short-term market noise and incentives and stay focused on long-term objectives, thus increasing the company’s value for all shareholders.
While I agree with much of what he says, I want to make a few objections.
- Ignoring short-term market signals isn’t necessarily a good thing. Does the market exhibit Brownian motion? Sure. Often, though, it’s telling you that you’re going in the wrong direction. E.g., the market has effectively said that Yahoo’s shareholders would have been better off taking MSFT’s bid. Yahoo has ignored that statement, likely to its’ shareholders’ detriment.
- Marc is implicitly asserting that the founders will have a better sense for the long-term direction than will the market - and that assunmption is often going to be wrong. First of all, the wisdom of founders can vary pretty widely. Second, even if the founders are pretty smart, there are cases when the market is smarter. Jerry’s a sharp guy. Did he make the right choice over the last few weeks? The market says “no”, and my instincts agree.
These two points can end up being semi-religious discussions about market efficiency, blah blah blah. I tend to believe in market efficiency - or, at least, the notion that the market is smarter than most people, most of the time - but it’s a reasonable debate, and one could also suggest that short-term market noise & incentives creates some sort of “friction” that destroys value for typical companies, even if the pricing signals are accurate on average. In any case, however, my biggest concern with Marc’s argument is this: that goals are, very often, not terribly well-aligned. Marc correctly notes that alignment of goals is key, and includes at the end a set of conditions for dual-class structures to work effectively:
The third point is the one that seems quite suspect to me. Even if they’re in the same boat from a share-price perspective, their goals can be quite different. Founders who are already rich may be driven by ego - to take a longshot at creating the most valuable company ever, to enjoy the power of running a big organization, to be in the press all the time - or other factors, that may at times indicate directions quite different from those that would maximize economic return to shareholders. (Founders who aren’t rich might have a strong incentive to cash out quickly and retire, when building a long-term company might be optimal.)
"Treating all other shareholders fairly" is a nice principle, but in the end, the controlling voters will maximize their interests, and share-price alignment doesn’t mean that overall goals are truly aligned.
This doesn’t mean that I oppose dual-class structures, by the way. If I were an entrepreneur and I could get such a structure (and still raise enough capital), I’d do it too (that way, I get to maximize my own interests, right?). … and if, as Marc says, there’s no bait-and-switch - i.e., all investors buy in knowing the structure - then there’s no ethical problem. The question is whether or when it’s a superior structure for maximizing the value of the company, and that’s where my concerns above apply.