Top IPO, Weak Governance

According to media reports, in what is expected to be the largest IPO of all time, SpaceX is seeking to raise as much as $75 billion at a valuation of more than $2 trillion. SpaceX could well have assets with very high value and exceptional growth prospects, and investor enthusiasm about it might thus be fully understandable. However, SpaceX also has poor governance arrangements which would have considerable adverse effects on public investors.

In this post, we examine some key governance defects in the SpaceX structure. We explain how they should be expected to (i) provide Musk with substantial value at the expense of public investors and (ii) produce value-decreasing inefficiencies by distorting incentives and decisions. Public investors seeking to determine the price at which they would be willing to purchase SpaceX shares should recognize and take into account these governance flaws.

(The prospectus of SpaceX was not publicly available as of the time that this post was written; the discussion below is based on media reports based on reviews of the draft of the prospectus.)

In particular, we discuss below in turn:

(a)The risk that Musk would over time become a controller with a small-minority or even very-small-minority stake which would involve a poor alignment of his and public investors’ interests;

(b)The perpetual nature of the small-minority control structure, which  should be expected to remain in place even if this structure proves highly inefficient; and

(c) Why Musk may have had incentives to adopt a governance structure that provides him with private benefits even if it is substantially inefficient.

The Perils of a Small-Minority Controller

The existing governance structure of SpaceX would down the road enable, and likely lead to, a governance structure with a small-minority controller. The term “small-minority controller” was put forward in our article The Perils of Small-Minority Controllers (Bebchuk-Kastiel (2019)), which provided a detailed analysis of the distortions and inefficiencies produced by such structures.

At its IPO, SpaceX will sell Class A shares to the public, each with one vote per share. It also has super-voting Class B shares, each with ten votes per share. Musk currently owns about 40% of the company’s equity capital while controlling about 80% of the voting rights.

Over time, the fraction of B shares that will be held by Musk should only be expected to increase for several reasons. First, any B shares held by non-Musk shareholders will be automatically converted to low-voting shares if sold to non-Musk entities; B shares that will be sold or transferred will retain their superior voting rights only if the sale or transfer is to Musk or Musk-related entities. Second, the company will be able to issue new B shares only to Musk and Musk-related entities, and the company should be expected to provide Musk with additional B shares as part of his compensation arrangements.

Note that non-Musk holders of B shares at the time of the IPO would have no reason to prefer those shares over A shares. With Musk having a lock on control in any event, the number of votes these holders possess will have no practical effect on their interests. Over time, these holders may decide to cash out, in which case the sold shares would lose their superior voting rights. Furthermore, if Musk led SpaceX to offer other Class B holders an opportunity to exchange their B shares for A shares at, say, 1.1 A shares for each B share, all non-Musk holders of B shares would likely use this opportunity.

The dual-class structure discussed above implies that Musk would be able to reduce considerably his equity stake at SpaceX without weakening his grip on control. If SpaceX goes public with a valuation of $1.75 trillion, SpaceX shares would likely represent a significant majority of Musk’s wealth. Standard risk diversification will provide Musk with substantial incentives to unload some of his shares and reduce the size of his ownership stake.

How much would Musk be able to reduce his ownership stake and still have a lock on control? Musk would be able to sell all of his A shares and then as many of his B shares (which would convert into A shares upon sale) as possible without falling below 50% of the voting power. Assuming this occurs after non-Musk holders of B shares have sold or exchanged their shares for A shares, it would be sufficient for Musk to retain a number of B shares exceeding 10% of the outstanding A shares. In that scenario, Musk would hold about 9.1% of the company’s equity capital while maintaining an absolute lock on control.

It is worth noting that, if he chose to do so, Musk could retain his lock on control while further reducing his equity stake. As we explain in our recent article Controllers Unbound, given the currently lax state law constraints on controlling shareholders, Musk could introduce nonvoting shares and use them for this purpose. His control would enable him to arrange for SpaceX to issue a large number of nonvoting shares and distribute them to shareholders pro rata. Suppose SpaceX distributes to shareholders two nonvoting shares for each A share or B share held. Following such a distribution, Musk could sell all of his nonvoting shares –cashing out two-third of his equity stake in the company – without reducing in any way the number of votes he would hold by virtue of his B shares.

The potential emergence of a small-minority controller or a very-small-minority controller poses substantial governance risks because such structures could well produce costly distortions and inefficiencies with respect to a wide array of corporate decisions. This would be the case, for example, with decisions regarding related party transactions with Musk-affiliated entities, allocating investment opportunities between the company and Musk-affiliated entities, and the design of Musk’s compensation arrangements.

Suppose SpaceX would have to decide whether to take an action that would provide Musk a private benefit of B but would decrease the cash flows of all shareholders by L. Assume Musk holds a fraction α of the company’s shares. In this case, Musk would bear a share α of the loss L. However, Musk would still prefer to take this action as long as B – αL is positive, which is the case as long as L < B/α.

Note that the magnitude of this problem increases as the size of α decreases. As is consistent with the empirical work on this subject, a decrease in the fraction of equity capital held by a controller is associated with higher agency costs and lower company value

Thus, while fully accepting that Musk is an exceptional visionary and business leader and that SpaceX shareholders are fortunate to have him at the helm at present, shareholders should recognize that the company’s governance structure would likely produce value-decreasing distortions (compared with a scenario in which Musk led SpaceX to an IPO with a single-class, one-share-one-vote structure).

The Perils of Perpetual Control

Even fully accepting that Musk is currently by far the best person to have at the helm, this might not be the case forever. Therefore, it is important for SpaceX shareholders to recognize that the governance structure would not only provide Musk with a lock on control that would not be weakened by any reduction in Musk’s fraction of equity capital, but that this structure should be expected to last forever. (Our understanding is that SpaceX is not expected to go public with any sunset or other charter provisions that would eliminate or weaken Musk’s lock on control down the road.)

Business history teaches us that business leaders who are exceptional at one point in time often cease to be so over time. Will Musk still be the best leader for SpaceX in, say, thirty years, when he is 84 rather than 54? Even his most fervent admirers  should rationally recognize that there is a substantial risk that he will not be. Moreover, under a dual-class structure with perpetual control, shareholders should recognize that if Musk passes control to one of his children, this heir might not be up to the task.

In The Untenable Case for Perpetual Dual-Class Structure, we analyze the costs of an indefinite, perpetual dual-class structure. We explain that, no matter how well a founder leads the company at the time it goes public, there is a growing risk over time that this would cease to be the case.

It might be argued that this risk should not be given much weight by shareholders because, if a controller ceases to be a fitting leader for the company, the controller could benefit by relinquishing control and passing the helm to a fitting leader. However, this risk should not be dismissed. To begin, Musk (or a heir of his) might fail to recognize when their abilities are not up to snuff. Furthermore, because a majority of the costs of an ill-fitting leader would be borne by public investors, Musk (or a heir) might prefer to stay at the helm and continue to enjoy the private benefits of control coming with it. (Recall that Sumner Redstone, the small-minority controller of Viacom, remained in control in his nineties even though he reportedly could not speak, stand, or write clearly.)

Does Musk Have Incentives to Avoid an Inferior Governance Structure?

The analysis above might be opposed on grounds that Musk has an incentive to avoid a poor value-decreasing governance structure because any defects of this structure would be reflected in the price that public investors would be willing to pay for shares at the IPO. Because founders who take companies public internalize some of the costs that poor governance would subsequently impose on public investors, some financial economists hold the view that IPO structures should be presumed to be efficiently chosen (this is the so-called Jensen-Meckling (1976) logic).

However, in this case, Musk might well have incentives to prefer a governance structure that would increase his private benefits of control even if this structure would be overall value-reducing. Recall that according to media account Musk owns about 40% of the equity capital prior to the company’s going public. As a result, he would internalize only part of the adverse effects that an inferior governance structure would be expected to impose on public investors.

Suppose that, compared with going public with a one-share-one-vote structure in which Musk initially has control over SpaceX but relinquishes it over time, the governance structure chosen for the SpaceX IPO should be expected (i) to provide Musk with private benefits of $100 million via, say, related-party transactions, mega pay arrangements, and allocation of opportunities to Musk-affiliated entities, but (ii) reduce the value of cash flows that would be shared by all shareholders by $200 million. In this case, Musk would prefer to have this governance structure because he would capture fully the value of the $100 million in private benefits but would bear only 40% of the $200 million loss (the 60% remainder of $120 million would be borne by other pre-IPO shareholders).

===========

To conclude, an examination of the quality of the SpaceX governance is worthwhile even for public investors who have an extremely favorable view of Musk’s leadership talents and who believe that founders often have strong incentives to design the IPO charter efficiently. When such an examination is undertaken, it indicates that, even though SpaceX is a top IPO, it likely offers an inferior governance structure.

Note: In the interest of full disclosure, we note that Lucian Bebchuk served as an expert in the Tornetta case involving Musk’s Tesla compensation.