The number of private firms going public in the U.S. has declined significantly since 2000. A related phenomenon is that most private firms that “exit” (change ownership structures to allow early equity investors such as entrepreneurs and venture capitalists to cash out) choose to be acquired by another firm. In a new paper, we aim to provide insights into the above trends by empirically analyzing two related research questions using a comprehensive dataset on private firms from the U.S. Census Bureau. First, what explains the tremendous decline in IPOs in the U.S. since 2000? Second, what drives the dramatic shift toward acquisitions rather than IPOs in the case of exiting private firms since 2000?

Specifically, we test the following five (not mutually exclusive) hypotheses motivated by the theoretical literature or common beliefs:

• Hypothesis 1 “Weaker economy:” The number of private firms that can exit successfully (either through an IPO or an acquisition) went down significantly after the year 2000 relative to the pre-2000 level.
• Hypothesis 2 “Greater sensitivity to product market competition:” A standalone public firm is more prone to product market competition in the post-2000 period than in the pre-2000 period, so that a greater fraction of exiting private firms would choose to be acquired rather than go public. These exiting firms are not strong enough to sustain the increased product market rivalry post-2000 as standalone public firms, but can survive by selling to other companies, which can help the exiting firms on the product markets (Bayar and Chemmanur (2011)).
• Hypothesis 3 “More private equity financing:” The greater supply of private equity financing in the post-2000 period may have led to the decline in IPOs. For example, Ewens and Farre-Mensa (2020) suggest that the deregulation in securities laws in late 1990s made private equity financing more abundant to late-stage entrepreneurial firms in the post-2000 period.
• Hypothesis 4 “Smaller net financial benefits from being a standalone public firm:” The financial benefits of going public (arising from the lower information asymmetry and therefore greater stock liquidity) may have declined or at least remained the same after early 2000s. For example, some have argued that the financial benefits declined partly due to a decrease in the number of sell-side analysts after 2002 (see, e.g., Gao, Ritter, and Zhu (2013)). In the meantime, the additional regulatory requirements imposed on public firms in the early 2000s (namely, Regulation Fair Disclosure (Reg FD), the Sarbanes-Oxley Act (SOX), and the Global Analyst Research Settlement]) may have made going public more expensive. These regulatory changes have been widely blamed as the cause of declining IPOs (see, e.g., Zweig (2010), Weild (2011)).
• Hypothesis 5 “Increased need for confidentiality:” The importance of intangible assets has gone up significantly starting in the early 2000s, which, coupled with the unavoidable release of confidential information at the time of and subsequent to IPO, implies that a greater fraction of U.S. private firms, especially those concerned with leaking valuable information to competitors, will choose to remain private or delay going public to the extent possible. We do not expect a similar effect on private firms’ exiting through acquisitions.

Existing literature has examined several of the above explanations of the disappearing IPO phenomenon, but many of these studies to date conduct analyses using firms that have already gone public. In contrast, we use proprietary U.S. Census data, namely, the Longitudinal Business Database (LBD), the Census of Manufacturing Firms (CMF), and the Annual Survey of Manufacturers (ASM), to conduct a comprehensive analysis of the decline in IPO volume and entrepreneurial firms’ relative tendency to exit through acquisitions and thus provide new evidence on the disappearing-IPO puzzle.

First, consistent with the existing literature, we observe a significant decline in the number of IPOs since 2000, even after controlling for the changing characteristics of private firms in the U.S. economy. Interestingly, we find that small firms do not experience a larger decline in IPO propensity than large firms. Thus, the evidence using private firms contradicts the conventional wisdom that the puzzle of disappearing IPOs is mainly attributable to the decline in IPO propensity among small firms.

Next, we find that the number of private firms, the fraction of high-quality (eligible to exit) private firms, and the average quality (in terms of total factor productivity (TFP), sales, or employment) of private firms increase from the pre-2000 period to the post-2000 period. These results do not support the weaker economy hypothesis.

The greater sensitivity to product market competition hypothesis and the more private equity financing hypothesis both predict that the quality threshold of going public becomes higher in the post-2000 period than in the pre-2000 period, because only higher-quality firms can fend off the greater product market threat or need additional public financing in the presence of more abundant private equity after 2000. In contrast, neither hypotheses clearly predicts the quality threshold of being acquired to be higher in the post-2000 period. We find evidence consistent with these predictions. Specifically, we find that the differences in quality (in terms of TFP and sales) between IPO firms and acquired private firms and those between IPO firms and remaining-private firms have increased after the year 2000. Additionally, multivariate analyses based on different test designs consistently show that firms with higher TFP are more likely to go public relative to remaining private in the post-2000 period than in the pre-2000 period, but they are not more likely to get acquired after the year 2000. Further consistent with the greater sensitivity to product market competition hypothesis, firms in more competitive industries and those with fewer business segments are less likely to go public after the year 2000.

We also find additional evidence in support of the more private equity financing hypothesis. Specifically, firms in states or industries with greater venture capital (VC) investments experienced a larger decline in IPO propensity than their peers. Moreover, the IPO firms backed by venture capital have significantly lower post-exit long-term TFP than matched (similar) private firms that are also VC-backed in the post-2000 era relative to the pre-2000 era, while this pattern is absent among IPO and matched private firms without VC backing. This evidence suggests that the marginal benefit of going public (and raising capital from the public market) relative to staying private (and raising capital from private equity financing) is lower in the post-2000 era.

We find mixed evidence for the other two hypotheses. Consistent with the smaller net financial benefits hypothesis, we find that firms in industries with less analyst coverage (and thus smaller financial benefit of becoming liquidly traded public firms) are less likely to go public after the year 2000. However, there does not seem to be a significant change in IPO propensity around early-2000s’ regulatory changes that substantially increase the financial costs of standalone public firms, which runs against this hypothesis. Finally, by examining the exit choices of high-tech firms, which might be more concerned about confidentiality, we fail to find consistent evidence for the increased need for confidentiality hypothesis.

In all, our findings shed new light on the puzzle of disappearing IPOs as well as the growing propensity of entrepreneurial firms to exit through acquisitions. Using proprietary micro-level data on private firms, we provide a comprehensive picture of the disappearing IPO phenomenon in the post-2000 period and show that this puzzle is driven by multiple underlying economic factors, including the evolving product market dynamics and the increased supply of private equity financing. As long as these economic factors persist, we probably will not see a return of IPOs to their pre-2000 level, even after the Jumpstart Our Business Startups Act (JOBS Act) in 2012 and other recent U.S. legislative efforts to revive the IPO market.

This post comes to us from Thomas J. Chemmanur at Boston College, Jie (Jack) He at the University of Georgia, Xiao (Shaun) Ren at the University of Georgia, and Tao Shu at the Chinese University of Hong Kong, Shenzhen and Shenzhen Finance Institute. It is based on their recent article, “The Disappearing IPO Puzzle: New Insights from Proprietary U.S. Census Data on Private Firms”, available here.