Bo Bian is an Assistant Professor in Finance at the University of British Columbia; Yingxiang Li is a PhD Candidate in Finance at the University of British Columbia; and Casimiro A. Nigro is an Assistant Professor in Law and Finance at Goethe University. This post is based on their recent paper and is part of the Delaware law series; links to other posts in the series are available here.
Introduction
In 2013 the Delaware Court of Chancery’s came to a final decision regarding the by-now-famous Trados case. Trados involved claims against the board of a startup company that was sold in a merger transaction. Plaintiffs, who held common stock of the company, alleged that board members affiliated with the company’s VC investors were conflicted in approving the transaction. The VC investors held preferred stock that provided for a “liquidation preference” in the event of a company sale. Because of that liquidation preference, the VC investors received all the merger consideration while common shareholders received nothing. In an initial opinion, the court denied defendants summary judgment and determined that the claims should be evaluated under the plaintiff-friendly fairness standard. (See In Re Trados Inc. Shareholder Litigation, 2009 WL 2225958 (Del Ch. 2009.)) In a subsequent trial court opinion, the court confirmed the applicability of the fairness standard, but ultimately ruled in favor of the defendants because the common stock were likely to have no value – making zero a fair price. (See In re Trados Incorporated Shareholder Litigation, 73 A.3d 17 (Del. Ch. 2013.)
Trados has sparked vivid debate among academics and practitioners as well as within the PE (i.e., private equity) and VC (i.e., venture capital) industry. In particular, academics have discussed Trados along several dimensions: Qualitative analysis posits that Trados is of limited significance (Cable, 2020). A competing piece in theoretical law & economics, on the other hand, suggests that Trados imposes significant costs on VC. The authors add that though not without merit, the arguments in favor of Trados are not sufficiently general or compelling to justify the costs of the new regime (Sanga & Talley, 2021).
Our paper adds another piece to the puzzle by providing empirical evidence of the consequences that Trados has had on the “fire sales” of VC-backed start-ups in this changing corporate law environment (although we refrain from an assessment of its efficiency or inefficiency).
Background Information
Acquisitions of VC-backed firms are often at the center of the controversial litigation that has its root in both the financial structure and the governance model of VC-backed firms. VC-backed firms typically issue two classes of shares: common shares for the entrepreneurial team and preferred shares for VCs. The diverging payoff functions of preferred and common shares can lead to well-known conflicts of interests between common and preferred shareholders. In addition, VC-backed firms’ boards often have members that are so-called “dual fiduciaries.” On the one hand, VC-appointed board members have powerful financial incentives to maximize value for ultimate investors in the VC fund. To the extent that they act as GPs of the VC fund, they also share in the legal obligation to do so. On the other hand, the VC-backed company’s board is under the fiduciary duty to maximize the value of the corporation for the benefit of its common shareholders.
The problems stemming from a VC-backed firm’s financial structure and governance model become more prominent due to the key features of the VC business model. VC funds have a finite lifespan of up to 12 years. They face mounting pressure to liquidate as towards the end of their lifespan. Portfolio companies’ boards – particularly VC-appointed board members – may thus push for “rushed” acquisitions that achieve lower sale prices and satisfy VCs’ liquidation preferences but leave little-to-nothing for common shareholders, who may therefore search for relief through litigation.
The Empirical Findings
Our work studies the “fire sales” of VC-backed start-ups before and after Trados. Before Trados, Delaware law was friendlier to preferred shareholders – allowing a VC-controlled board to make decisions that favored the preferred shareholders at the expense of common shareholders. After Trados, all else equal, common shareholders are able to credibly threaten directors with fiduciary duty litigation. In effect, Trados has empowered common shareholders to hold-up a sale, thus affecting the probability of VC exits through acquisitions, especially when VCs move closer to maturity and experience greater pressure to liquidate their investments.
We begin by identifying potential “forced sales” in the VC industry. Recall that VC funds have a limited lifespan of up to 12 years, which generates mounting pressure to liquidate as the end of their lifespan approaches. Consistent with this, we observe that the median VC fund has distributed 93% (100%) of its cumulative cash by age 12 (15). We thus define forced sales as acquisitions that take place when the fund is close to age 12, with a window that spans from age 11 to 13. Using detailed transaction data, we provide strong evidence of fire sales in the VC setting. First, we find that forced sales achieve a lower deal price. In particular, for sales within our window, the sale price is almost 30% lower than for deals closed in other years. We establish our benchmark price using variables that are known to predict investment outcomes such as the total amount of equity raised, the number of financing rounds, and the VCs’ selection skills. Moreover, we find that the discount in sale price increases with the urgency of the sale. Second, we find that these VC-backed start-ups are more likely to be acquired by industry outsiders, including financial firms and, most notably, private equity investors. These industry outsiders face significant costs when acquiring and managing target assets, driving down their valuation – a potential explanation for the fire sale discount observed above. Third, we detect a higher-than-normal return earned by acquirers in transactions affected by VC funds’ liquidity pressure. This abnormal return also increases with the fire sale discount, suggesting that part of the target’s loss becomes the acquirer’s gain.
Next, we seek to understand whether and how the shock caused by Trados in terms of improved common shareholder protection affects the timing of VCs’ acquisition exits. To this end, we deploy a difference-in-differences (DiD) estimation method and compare the probability of exits through sales by VCs nearing maturity versus VCs further away from maturity both before and after Trados. Consistent with the notion that Trados gives common shareholders more leverage to challenge the sales of VC-backed companies, we find that VC funds now act more cautiously in exit decisions, especially when they are under liquidity pressure. Since Trados, maturing VC funds are less likely to exit through sales that are under-priced and costly for common shareholders. In fact, Trados mitigates most of the positive effect of liquidity pressure on the probability of exits through acquisition in scenarios which typically feature the most substantial liquidity pressure (and therefore the most extreme fire sales). Trados still has a significant, albeit considerably smaller, mitigating effect when the sales are less costly to common shareholders
Since Trados is a Delaware opinion, we also examine whether the treatment effect is indeed driven by start-ups incorporated in Delaware, to which Trados applies most directly. We find evidence consistent with this hypothesis, alleviating the concern that coinciding events or other general shocks in the VC sector may drive our findings.
We conduct an additional test of our hypothesis by analyzing distributions to LPs: If VCs are less likely to exit through rushed acquisitions near the end of the conventional fund lifespan, their proceeds and cash distributions to LPs should be affected accordingly. Using a similar DiD setting, we investigate fund distribution patterns before and after Trados. We find that VC funds that face high liquidity pressure are less likely to distribute cash to their LPs since Trados. Cash distributions to LPs towards the end of the conventional VC lifespan accordingly make up a lower proportion of total distributions since Trados.
As a final step, we examine the effect of Trados on ex-ante VC fundraising and start-up financing. Moving from a preferred- to a common-favoring regime may discourage LPs to invest, reducing their allocation to VC funds. We find that this is especially true for foreign LPs. After all, one of the primary motives behind investment in foreign markets such as the US is the exploitation of investor-friendly legal environments. We accordingly document a decrease in the size of US VC funds relative to non-US funds, which suggests the decreased availability of VC for US start-ups. By examining worldwide VC deals, we also find a reduction in capital raised by companies located in the US. Taken together, these results suggest that while Trados indeed reduces the likelihood of VC exits through fire sales, these benefits for VC-backed companies and common shareholders come at the cost of a reduction in the total supply of VC.
Implications (in Brief)
Overall, our results highlight some of the important trade-offs in contemporary corporate law-making and offer valuable guidance to lawmakers and practitioners in the VC and start-up community.
The complete paper is available for download here.