What’s Wrong with the IPO Process and How to Fix It

While DoorDash and Airbnb successfully completed their much-anticipated IPOs this past week, there has been a lot of scrutiny on how these IPOs were priced and subsequently traded up significantly. The stock price for DoorDash rose 86% at the start of trading, and Airbnb went a whopping 115% above its IPO price; in both cases the companies could have raised significantly more if they were priced more closely to the levels at which they traded.

The trend of IPO pricing and then “popping” is likely to continue, with several closely watched IPOs potentially coming up in 2021 – including Instacart, Robinhood, Stripe, and SpaceX. David Erickson in this opinion piece explores why this disconnect between pricing and trading happens and how it could be fixed. He is a senior fellow in finance at Wharton, where he co-teaches “Strategic Equity Finance” and is co-director of the Stevens Center for Innovation in Finance. Prior to teaching at Wharton, Erickson worked on Wall Street for over 25 years, helping private and public companies raise equity strategically.

After two of the largest and most high-profile IPOs of the year, DoorDash and Airbnb, priced and then “popped” significantly, there was a lot of focus on what went wrong with the pricing — from this Bloomberg article to Jim Cramer on CNBC calling IPO pricing “broken” and “embarrassing.” This isn’t the first time this scrutiny has happened, and it probably won’t be the last time — unless there are changes made to the IPO process.

A few years ago, I raised some of the issues that create this disconnect between pricing and trading in a Knowledge@Wharton piece titled “What’s Wrong with the IPO Market? That is the Wrong Question.” While there have been a couple of changes in the last few years in the Going Public process from the emergence of direct listings by a few “unicorns” (led by Spotify in 2018) and the evolution of the SPAC merger market this year (with over $125 billion in SPAC mergers announced already this year, according to Axios), the IPO process will likely continue to be the way most private companies access the public markets.

Why Does the Significant IPO Pricing/Trading Disparity Happen?

While the “pop” doesn’t happen on most IPOs, a few factors have probably driven this craziness in last week’s DoorDash and Airbnb IPOs to levels we haven’t seen since the 1999/2000 “tech bubble”:

NASDAQ is up almost 40% YTD;

“Cult” stocks like Tesla (up almost 630% YTD to a $568 billion market cap at this writing) have captured the imagination of retail investors (often referred to as “Robinhood traders” in 2020); and

Recent high-profile IPOs like Snowflake (closing up 112% on its first day, after raising its IPO price by 50% since initially filing) and BigCommerce (closing up 201% on its first day) had big “pops.”

Why does this happen?

Having been involved in pricing many IPOs when I was on Wall Street (and now following them for almost eight years while at Wharton), the three primary reasons for this disparity for high profile IPOs surprisingly remain the same:

Limited Supply – Many high-growth companies sell a little bit at IPO (sometimes less than 10% of the company) with the hope they can potentially sell more at higher prices as the stock “seasons.” As an example, if a company sells 10 million shares at IPO (out of its 100 million shares outstanding post-IPO), and it is a “hot” company, you may see in the media (while the IPO roadshow is happening) that the offering is significantly over-subscribed. Hypothetically, if the offering is, say, 30x over-subscribed and thus 300 million shares of demand, while the company can often increase the size of the IPO, they often still only sell 10 million shares. As a result, there is 290 million shares of demand, much of which could be bought once the stock starts trading. It should also be noted that the 300 million shares of demand is not the total potential demand for the IPO, but it is the demand that is received from the institutional and retail clients of the banks that underwrite the IPO transaction.

Significant Retail Demand – Beyond the 290 million shares of unsatisfied demand from the IPO, in this hypothetical deal, there can often also be a significant amount of interest from both institutional and retail investors that didn’t have access to the IPO. Often, this develops for high-profile consumer/retail-recognized companies (like DoorDash and Airbnb) doing their IPO, and/or IPOs that benefit from media coverage of IPO companies that have significant momentum during their roadshow (like Snowflake, which raised its IPO pricing range multiple times during the two-week roadshow process). Unfortunately, none of this demand can be captured and quantified by the IPO underwriters, who due to required Security and Exchange Commission (SEC) filings are usually pricing the IPO the night before it starts trading (and this demand is visible).

Which leads me to the biggest reason I believe this pricing/trading disparity exists:

Time Disconnect Between SEC IPO Filing Process and IPO Trading Process – As I mentioned above, the timing of required SEC filings can exacerbate the pricing/trading disparity in two big ways:

During the roadshow – Because of potential demand, if the company and underwriters believe it is possible to raise the size of the offering (usually by more than 20% from where the “Red Herring” IPO marketing document is filed), the company would need to file an amendment to its registration statement with the SEC revising the range/size. Because this is a public filing, it is usually picked up and reported by the media, alerting potential investors that there is significant demand for the transaction. As an example, Snowflake filed its “Red Herring” document on September 8 (as the roadshow was to start) with a price range of $75 to $85 per share and amended its filing on September 14 (one day before pricing) with a price range of $100 to $110 per share.

On the night of pricing (the day before trading) – To continue this example, Snowflake was declared effective on the evening of September 15 (the night of pricing), announcing its price of $120 per share – 50% higher than when the IPO roadshow started!

As each of these filing events with the SEC are public, the media often reports them. This coverage can often increase the momentum and the interest in the transaction as investors realize that the stock is likely to “pop.” Given all of this interest and demand on the morning of September 16, Snowflake’s stock was unable to be opened until early afternoon and finally opened at $245 per share (more than doubling from pricing).

Why does the SEC do this? To protect investors with the updated disclosure information, including in Snowflake’s case how a change in proceeds raised alters the numbers that “flow through” the document (i.e., changes in Summary Financials, Use of Proceeds, etc.).

How Can the IPO Process Be Fixed?

While Direct Listings and SPAC (Special Purpose Acquisition Company) mergers have been offered as alternative solutions for private companies to go public, the right answer for most companies is to fix the IPO Process, as it will likely continue to be the primary way most will go public. Here are a few ways the process can be fixed to make it a more efficient and effective process (but they won’t be easy):

Streamline the IPO Process – From the time a company initially files its registration statement with the SEC to when the company can price its IPO, it usually takes a few months. Despite all the technological developments over the years, this timeline surprisingly has not changed much since way back when I started on Wall Street. With the pandemic this year, the traditional IPO roadshow, where the company’s management team historically would fly to meet with major institutional investors across the country over a seven-to-ten-day period, obviously had to be changed to Zoom meetings. While I assume going forward the roadshow will be much more streamlined, the biggest timeline issue is the “back and forth” with the SEC on the registration statement (and subsequent required amendments), which is usually at least six to eight weeks. It seems like this process could be more efficient (and less costly) for companies, but still achieve the level of disclosure the SEC requires for investors.

Improve the Transparency with the Overall Demand – Not just the demand from the IPO roadshow, but also the demand from those that didn’t have access to the roadshow or who aren’t clients of the IPO underwriters (i.e., all of the demand as the IPO starts trading). The challenge for those companies that aren’t part of the IPO underwriting/selling group (i.e., Robinhood, other retail systems): What is their incentive to show their demand unless they either get access to IPO shares or get paid for this information in some form?

Collapse the Time Disconnect Between the SEC Filing Process and When the Stock Trades – This is the biggest potential fix, but also the most challenging. If the IPO underwriters can wait to price the deal until the stock is close to opening, it can maximize the price for the issuer and minimize pricing inefficiencies. While this sounds great, it probably requires the SEC to be more flexible on the timeliness of some disclosures — if it significantly changes the “makeup” transaction other than increasing the size (i.e., if the company is selling less shares, and shareholders — including management — are selling more as a percentage of the offering) it needs to be disclosed the night before as with the current process; if it is just a matter of price, it can be disclosed as the stock is to open. The other big challenge to recognize is that while this fix may help the minority of “hot” IPOs like DoorDash and Airbnb that significantly traded up, it will potentially create more difficulties for the many IPOs that may struggle in their debut. Do not forget just over a year ago, both Uber and Peloton were in the latter group, having struggled below their IPO price in their respective debuts.