IPOs Since the 1920s

Companies conducting an initial public offering of their shares are often newer and less likely to be profitable, or at least consistently profitable, than long standing companies. There may also be less information available about them. So, arriving at an appropriate price for these shares and buying them is riskier than doing the same for shares in other companies. At least today, there is substantial regulation of IPOs which among other things require various disclosures about the issuer. It is plausible to assume this has improved IPO pricing. However, how exactly IPOs are priced is not obvious, straightforward, or consistently friendly to share issuers.

Initial Public Offerings

In the 1920s, American IPOs bore substantial resemblance to those of today. Investment banks underwrote new stock issues and partnered with brokers to syndicate the shares. The investment banks also agreed to support the share price during the offering period which was crucial to keep people investing in the shares. New offerings were a slightly more drawn-out process in the 1920s and if the share price fell after being first issued, it would imperil the successful conclusion of the offering.

New York Stock Exchange, 1923 (source: geographicguide.com)

Along this line, IPOs in the 1920s were different in that shares would often not be listed on a proper exchange until a few weeks or months after the IPO, though these securities did trade over-the-counter and on the less formal New York Curb Exchange in the meantime. Today, these events are largely simultaneous. Nonetheless, the support bankers provide to the share price in the aftermath of an IPO is a vestige of this slower process.

Bankers and share-selling syndicates earned fees for distributing the shares to investors. For the brokers selling shares to clients, they earned a commission. Some of the largest syndicates could comprise as many as nine hundred firms. Meanwhile, the lead bankers on the transaction earned extra fees for committing to hold unsold shares and organizing the offering. Banks often also received warrants in the companies they introduced to the market; these warrants gave them the option to buy additional shares at a fixed price.

Investment Trusts

The 1920s saw IPOs for ordinary companies as well as for financial vehicles called ‘investment trusts’. Unlike their British equivalents, American investment trusts were not mutual funds holding diverse investments for long time horizons. Rather, these were companies established to trade in shares often with limited disclosure to investors. They also borrowed money to add leverage to their trading activity. The investment trusts were underwritten by banks with great enthusiasm because in addition to earning their ordinary fees, these vehicles supported the IPO boom by buying shares in other newly-listed companies. An investment bank could make money underwriting newly listed companies and then the investment trusts that bought the shares in the former.

These investment trusts are not long-abandoned relics of a less well-regulated past in financial markets. They have been likened to the special-purpose acquisition companies (SPACs) which became commonplace new issuers in the early 2020s. In any case, investment trusts, which made up 30% of IPOs in 1929, were sponsored by investment banks and these banks often received warrants in the investment trusts they managed. The vehicles had high fees for end investors, often in the form of sales charges or ‘loads’, that could be as high as 30%. They also often traded at a premium over the value of their portfolio holdings. Some investment trusts invested in other investment trusts and this meant investors in these vehicles were buying into a very high fee structure at a high all-in premium.

Pricing

Pricing of IPOs has raised questions for decades, both before the introduction of new regulations for American IPOs in 1933 and since. It’s often curious that newly-issued shares almost invariably see large gains on their first trading day. After all, isn’t this appreciation a sign the IPO shares were underpriced to begin with and isn’t that a loss to the stock issuer. Wouldn’t underwriters be better at pricing IPO shares and market this advantage to win over new clients?

Regardless, there was some process used to price shares, however accurate or fair, or inaccurate and unfair, the final outcome was. Lead underwriters might consult large institutional investors in pricing a new security but members of the selling syndicate held larger sway. The latter’s input was crucial to properly pricing a new issue.

As for why IPOs appear underpriced, some argue that this is a form of financial insurance to the underwriter because these banks do not want to be stuck with shares they paid too much for. However, even IPOs where underwriters are not committing to hold unsold shares see large underpricing. Others argue that, especially for more speculative issuers, bankers have enough bargaining power to underprice issuers’ securities as an extra compensation to the bank and the investors. Another theory is that investment bankers had better information on the pricing of securities than the issuer and so could underprice the new shares due to an information asymmetry.

In the 1920s, IPO underpricing could be significant. Alleghany Corporation was an investment company underwritten by J.P. Morgan & Co., and it saw an IPO in 1929. The shares were offered at $20 per share and surged to $57 within five months. Nonetheless, this type of IPO was more speculative than normal.

During the 1920s, a typical IPO for a non-investment trust was usually for a fairly established and profitable firm. There was less uncertainty and more information available than might be commonly appreciated. It is true that disclosure requirements were lower but this was mitigated somewhat by the more established nature of your typical new share issuer and the more durable and perhaps closer relationship between a company and its principal bankers; the latter would know a lot about the former’s business.

Though the drivers of IPO underpricing are open to considerable debate, the statistics on IPO underpricing suggest the ‘problem’ has never been solved. In fact, it seems to have gotten worse. Excluding investment trusts, the average first-day return suggests IPOs were underpriced by 8.2%, on average, in the 1920s. Further, during the 1920s, it was extremely rare for a new issue to close lower on its first day. These numbers are for the decade as a whole. Still, the average underpricing in the peak years of the U.S. stock market boom, 1928 and 1929, was 7.4%, not meaningfully different from the average for the decade taken all together. As it happens, the average underpricing would be higher in later decades than these numbers from the 1920s.

Allocations

If investors know that IPO shares are underpriced, there is bound to be more demand for them than there are shares available. So, it’s perhaps inevitable that after underpricing, allocations are the second big mystery, and controversy, of IPOs. During the 1920s, investment banks like J.P. Morgan & Co. and Drexel & Co. maintained a ‘preferred list’ of clients who received priority in receiving newly-issued shares. Alleghany Corporation shares were allocated to politicians and financial executives in 1929. Amongst potential subscribers, banks had an incentive to allocate shares to those who did other business with the firm. Less distinguished customers had to wait to buy in, likely at higher prices.

Reform

Regulation of IPOs did increase with the 1930s New Deal. There were some legal restrictions on IPOs even in the 1920s though; these included state securities laws, mail fraud statutes, and protections against fraud offered by common law before the statues were reformed in the U.S. in the 1930s. Nonetheless, the Securities Act of 1933 mandated disclosures to investors and introduced registration requirements for new offerings. Underwriters’ fees must now be disclosed. There was also a cooling off period that allowed time for a security to be analyzed before it was sold. The later Securities Exchange Act of 1934 created the Securities and Exchange Commission and added new regulations for trading in shares that have already been issued, after the IPO.

Despite these reforms, underpricing of IPOs has lived on and even grew from 8.2% on average in the 1920s to 18.6% in the 1980-2020 period (Gan, Mahoney, Mei, 2022). During the internet bubble of the late 1990s, the average first-day return on technology stocks was 86.7%! Comparing the 1920s to IPOs in the late 1960s, as was done by Seha M. Tiniç (1988), also shows a larger underpricing in the post-Securities Act era than before.

Some of this increased underpricing may be a function of the typical new share offering of the 1920s being for an issuer that was older, on average, and more likely to be profitable than those of IPOs in the 1980-2020 period. Still, these factors alone do not seem to explain the difference when controlling for them in studies of underpricing. It has also been argued that the new regulatory scrutiny has encouraged IPOs to be more underpriced, not less, by creating a buffer protecting investors against potential losses which are now more likely to result in legal liability to the underwriter, especially for smaller more speculative new issues.

That said, there is one metric which suggests IPO returns were more artificially contrived in the 1920s than today. Negative first day returns for newly listed stocks are more common now than in the 1920s, when an IPO finishing its first day of trading at a loss was nearly unheard of, suggesting the pricing was perhaps too stable to be a genuine indication of market appetite in the 1920s. Surely, underwriters in the 1920s could not have been that good. Perhaps underwriters supported pricing of those first trades more thoroughly in the 1920s than today, making initial returns an unreliable proxy for IPO underpricing.

In any case, another controversy of the 1920s IPO boom, the role of investment trusts, remains relevant. This is because even the investment trust made a return of sorts with the special-purpose acquisition companies of the early 2020s. In their heyday, SPACs also traded a premium over asset value, 27% on average and as much as 40% in one case in March 2021, and these vehicles were also opaque to investors who would not know what they would invest into.

Lesson

Persistent IPO underpricing may seem inefficient if not abusive towards issuers looking to maximize the proceeds of their public offering. Similarly, ‘blank check companies’ like investment trusts and special-purpose acquisition companies garner suspicion. Yet, each has been around since at least the 1920s. Investors buying into these IPOs after the shares begin to trade will, on average, miss the easiest money. Those who receive allocations in a ‘hot issue’, sponsor a SPAC or similar vehicle, or distribute the shares realize their returns quickly, today just as a century ago.

More from the Tontine Coffee-House

Read about Ford’s 1956 IPO and how American regional stock exchanges, outside New York, differentiated themselves. Consider subscribing to this blog’s newsletter or checking out book recommendations, which include many of the sources often referenced in my posts.