Dual-class stock structures have proliferated in recent years. In 2017-2019, almost 30 percent of IPOs in the U.S. had a dual-class structure, and most of them were founder-controlled technology firms (Aggarwal, Eldar, Hochberg and Litov, 2020). Their increasing popularity has drawn the ire of many institutional investors, proxy advisory firms and academics who view dual-class structures as detrimental to shareholder value. Founders with superior voting rights may have strong incentives to extract private benefits or pursue fanciful projects at the expense of shareholder value.
The standard accounts of dual-class structures ignore the full menu of organizational choices that are available to entrepreneurial firms. The critique of dual-class structures rests on the assumption that the alternative to a dual-class IPO is a single-class IPO in which shareholders’ voting power is identical to their economic interests. However, if dual-class structures were not legally permissible, the firms that opt for dual-class structures could opt to remain private indefinitely or, at the very least, postpone the IPO.
The underlying motivation for adopting dual-class structures is that founders place very high value on maintaining control. If they cannot maintain control after the firm becomes public, they may choose not to become public at all. Indeed, during the same period that dual-class firms have proliferated, there has also been a dramatic increase in the number of startups with over $1 billion valuations, commonly known as unicorns. The proliferation of unicorns underscores that remaining private is a viable and attractive option for sizable startups.
From a governance perspective, the private option may be particularly concerning. Unicorns with dominant founders have experienced a multitude of scandals in recent years, some of which ended with major losses for investors. Well-known examples include the collapses of WeWork, Theranos, and FTX. These companies not only had a toxic work culture or failed to comply with laws and regulations, but more importantly, their business models were either fraudulent or could not realistically support the firms’ valuations. Most of these failures resulted directly from the irresponsible behavior of a dominant founder, and the inability of investors, including reputable VC (venture capital) firms, to monitor their actions in a material way.
In comparison with the dramatic governance failures in large unicorns, dual-class corporations have arguably fared reasonably well. While there is an ongoing debate about the performance of dual-class firms, overall performance appears to have been reasonably strong, at least when considering the founder-controlled firms that became public in the 2010s (see Ahn, Fisch, Patatoukas & Solomon, 2021). Moreover, none of them experienced a failure that amounted to a complete collapse of the business model.
The explanation I offer is that the IPO process is effective in distinguishing founder-controlled firms that have viable business models and valuations from those that do not. Perhaps the largest agency cost associated with founder control is that the founder may exaggerate the growth potential of the firm to raise capital at higher valuations. Investors can suffer extreme losses when they discover that the valuation is grossly inflated, or worse, that the business model is fraudulent, and the firm generates no value. By ensuring that the firm has a viable business model and a reasonable valuation, the IPO process mitigates the tail risk of the agency problem in founder-controlled firms that could result in dramatic losses for investors.
To the extent that they facilitate the IPO decision, dual-class structures effectively mitigate the agency costs of founder control. Without the option to create dual-class structures at the IPO, these founder-controlled firms may stay private, and if they do, they will escape the scrutiny of the IPO process. As the recent failures of large unicorns suggest, even reputable VC firms may fail to monitor startups effectively. The IPO process, which includes detailed disclosure and financial analysis, can elicit new information on these private firms that may end in the delay of their IPOs, adjustments to their valuations, or even the withdrawal of the IPOs.
In this sense, dual-class structures provide a solution to the unicorn governance problem. When there is great availability of private capital, founders can delay the IPO or keep their startups private. There is indeed evidence that tech unicorns that invest in intangible assets tend to go public later than other startups of similar age (Davydova, Fahlenbrach, Sanz & Stulz, 2022). The VCs who are scrambling to get a piece of a startup with substantial growth potential have little leverage in negotiating for control rights and often have very few tools to monitor the operation and even the strategy of the firm.
By acquiescing to the dual-class structure at the IPO, the VCs can get the startup founder to go public at a relatively early stage of the startup life cycle. Indeed, the average age of dual-class firms at the IPO is substantially lower than that of other IPO firms. By prompting startups to go public, VCs can reduce the risk of a major governance failure that they may be unable to prevent in an environment in which they compete for investments. Within this broader perspective that accounts for private markets, the dual-class structure is paradoxically a solution to the relative laxity in the governance of entrepreneurial startups.
This account provides an overlooked explanation for why VCs have warmed to dual-class structures, and some have even lauded them as an ideal structure for founder-controlled technology startups seeking to go public. The reason is that under economic conditions in which founders have the upper hand, VCs are likely to be less concerned about founders’ control after the IPO. Instead, their primary concern is that the firm may remain private without any meaningful scrutiny, thereby exacerbating the risk of a major failure.
This analysis underscores the role of the IPO process in instilling discipline within founder-controlled firms. The scrutiny of capital markets, combined with mandatory disclosure during the IPO stage, effectively filters out founder-controlled startups lacking viable business models. In equilibrium, startups lacking plausible models would opt not to pursue a public offering. The WeWork case may be perceived as an off-equilibrium event wherein the founder, Adam Neumann, chose to go public with Softbank’s support, seemingly expecting that public investors would overlook the facts disclosed in the company’s registration statement. The subsequent withdrawal of the IPO and the founder’s removal, prompted by weak demand from public investors, further emphasizes the disciplining impact of the IPO process.
It is important to emphasize that the accountability and transparency requirements that are imposed on public firms do not cure all governance problems, such as compliance failures or a toxic work environment (Platt, 2023). Many public firms, whether they are founder-controlled or not, have experienced such failures. While it is possible that being a publicly traded company reduces the risk of compliance failures and scandals, this appears to be largely a marginal impact of going public. Rather, it is the IPO process with its requirements for detailed disclosures about the firm’s business model and financial accounts followed by market scrutiny that is effective in screening the entrepreneurial startups that have good (or at least plausible) ideas from those that don’t. In this sense, my account of dual-class IPOs reveals the true role that being public serves in disciplining founder-controlled entrepreneurial firms.
My analysis of the role of dual-class IPOs might also have normative implications. In the absence of a dual-class structure, it’s conceivable that fewer unicorns would have opted for the IPO route. Consequently, without the prospect of an IPO, many unicorn founders might have experienced less oversight, attracting more private investment for potentially implausible or even fraudulent projects. The availability of the IPO option creates incentives for founders to avoid inflating startup valuations by making unrealistic promises. Paradoxically, the absence of dual-class structures could potentially lead to more pronounced and extreme startup failures.
Accordingly, in the enduring debate surrounding the advantages and drawbacks of dual-class structures, it is crucial to recognize their role in streamlining the IPO process for startup unicorns. When evaluating policies that seek to limit firms’ adoption of dual-class structures, the assessment should extend beyond their potential performance as single-class entities. Equally significant is an examination of their performance as private firms, taking into account the elevated risk of an extreme governance failure.
 Scott Kupor, the managing partner of Andreessen Horowitz, stated in 2013, “Dual-class stock is … well suited for founder-led technology companies …where economic interests between external shareholders and internal management are aligned.” See Scott Kupor, Sorry CalPERS, Dual Class Shares Are A Founder’s Best Friend, Forbes (May 14, 2013,) https://www.forbes.com/sites/ciocentral/2013/05/14/sorry-calpers-dual-class-shares-are-a-founders-best-friend/?sh=1ac94c1812d9.
This post comes to us from Professor Ofer Eldar at the University of California, Berkeley – School of Law. It is based on his recent article, “Dual-Class IPOs: A Solution to Unicorn Governance Failure,” available here. A version of this piece was originally posted on the ECGI blog on December 12, 2023,