Shares of Snowflake (SNOW -0.93%) recently debuted with a bang, closing its first day of trading 111% higher than its initial public offering price of $120 per share, and raising about $3.4 billion in the largest software IPO ever. For many institutional investors, such as hedge funds, pensions, and mutual funds, Snowflake's IPO is not one they will soon forget, as shares on the first day of trading hit $319 at one point. But for the average retail investor, it wasn't as much fun. The first trade on the public market opened at $245 per share and closed about 4% higher.

I myself ventured away from my traditional area of expertise to get in on the action, and ended up buying shares of Snowflake higher than I would have liked (so much for that huge POP). The experience begs the question of whether it's fair for institutional investors to enjoy a 111% bump, while the average investor only saw a mediocre 4% return after the first day of trading.


Image source: Getty Images.

How the IPO process works 

A traditional IPO is a long, tedious journey. The company, which will raise capital on the public market for the first time, hires an investment bank to serve as the intermediary between itself, the stock issuer, and the investors purchasing shares in a process known as underwriting.

The first big challenge for the company and underwriters is to set an initial range to price shares in. Because the company has never sold shares on the public market, this can be somewhat of a challenge. Usually the range is determined by looking at the company's valuation, which may be based on the amount of money raised in the private markets, revenue numbers, how competitors have been priced, the company's potential market size, and other attractive characteristics of the business model.

Next, the underwriters go on what is often referred to as a roadshow to drum up interest among institutional investors, many of which the underwriters already have a prior relationship with. This is an important part of the process because typically, institutional investors receive the bulk of shares during an IPO. Fidelity estimates the split to be about 90% to institutional investors and 10% retail, although it can vary . During the roadshow, underwriters are pitching the company going public and trying to gauge demand for the stock. The higher the amount of capital raised, the more in fees the underwriters can earn.

With Snowflake, underwriters appear to have been very successful on their roadshow. On Sept. 8, Snowflake updated its prospectus, pricing shares at $80, which the company said was the midpoint of the price range it had in mind. The filing also revealed that Berkshire Hathaway (BRK.A 0.23%) (BRK.B 0.27%) and the venture arm of Salesforce (CRM 0.64%) would both purchase shares in concurrent private placement offerings. This implied a higher market capitalization at the time and also likely signaled confidence to other investors, especially when you consider Berkshire Hathaway has never participated in an IPO before . Demand continued to build among institutional investors because Snowflake increased the IPO offering price to $120 per share on the day before the company went public on Sept. 16. 

On the morning of an IPO, the issuing company does not begin trading when the market opens at 9:30 a.m. EDT, because there is still a lot that needs to happen before the company makes its first public trade. The underwriters finally distribute shares to the institutional investors that expressed interest during the roadshow. In the Snowflake IPO, the underwriters also had the option to distribute an additional 4.2 million shares if the demand was there, in what is typically referred to as the "greenshoe" option. We know this happened with Snowflake . IPO Boutique reported that Snowflake's IPO was "many multiple times oversubscribed," which helps drive up the price .

After shares are distributed, price discovery begins. Orders to buy and sell the stock start coming in waves from retail investors and institutional investors, both ones that received the initial allotment of shares and participated in the IPO, and ones that didn't. According to Zacks, "buy and sell orders pile up until they are balanced against each other, determining the opening price. If the demand for shares exceeds the supply, the shares open higher than the offering price; otherwise they open lower ." 

Is it fair?

This is probably a question left up for debate, but there are a few things to consider. As mentioned above, retail investors do help drive the pricing during the discovery process. Snowflake, which was extremely hyped in the media, likely garnered a lot of attention from retail investors. Snowflake CEO Frank Slootman even told Fortune that he blamed some of the activity during the first day of trading on the "frothy, opportunistic retail side ."

Another thing to understand is that the reason the underwriters first sell to institutional investors is to create a liquid market. The founders, employees, directors, and some early investors who have shares going into the IPO are typically subject to lock-up agreements, which prohibit them from selling shares for a certain number of days. There were many articles following the IPO talking about how much money Berkshire Hathaway and Salesforce made on the first day of trading, but Buffett and Salesforce can't actually sell their shares for a year due to restrictions.

For the public to be able to buy shares, the underwriters need parties willing to sell shares. So, they incentivize institutional investors by choosing an offering price that will hopefully see enough of a gain on day one that some investors want to sell their shares. During the distribution of initial shares, the underwriters are careful to choose investors who will hold shares for the long term, and some who will flip the shares right away for the quick profit.

Another interesting point is that many believe the traditional IPO process through underwriters is actually unfair to the issuing companies. In Snowflake's case, the company, according to Fortune, sold 36.4 million shares in all. However, remember it sold those in the morning to institutional investors for $120 per share, raising a total of roughly $4.4 billion in cash (the $3.4 billion mentioned above does not account for the extra allotment and shares sold to Buffett and Salesforce).

But if Snowflake had decided to do a direct listing IPO, some argue that it could have raised much more than it did. A direct listing cuts out the process of hiring underwriters and selling shares to institutional investors, and hence a lot of the fees that come with that. Had Snowflake's stock price closed at $253.93 during a direct listing like it did on the first day of trading, Snowflake would have raised about $9.2 billion . "In many ways, $SNOW is the final proof of just how broken [the IPO] process is," Bill Gurley of the venture capital firm Benchmark tweeted following Snowflake's first day of trading . 

Slootman called this idea "nonsense" in an interview with Forbes. "We could've scraped a few more bucks off the table, but if you think you could have sold this offering at $245, I mean, people are smoking something. That is just ridiculous," he said. "This whole DL [direct listing] conversation is completely intellectually dishonest because of the notion of a demand curve." .

Is there a solution?

Obviously, many institutional investors are sitting on a huge profit right now, while most retail investors who bought on the first day are likely down or even, with Snowflake currently trading close to where it ended its first day. But it's important to note that retail investors did help drive the ultimate price at which the stock began trading, and that institutional investors do help create liquidity.

I also agree with Slootman in the sense that selling out Snowflake's IPO at $245 per share in a direct listing seems a bit rich. Additionally, Snowflake is the largest software IPO of all time, so you really can't expect to see this kind of demand from the retail side on your average IPO. Most companies likely need the institutional money to fill their orders.

That said, if platforms like Robinhood continue to increase the number of retail investors in the market, this is certainly an issue that should be studied. While a small pool of shares in an IPO can go to retail investors, I think it's mostly reserved for higher-net-worth individuals with hundreds of thousands to invest, instead of just thousands or hundreds.

Lastly, even though I'm holding shares of Snowflake at a loss right now, I can take solace in the fact that I didn't invest more than I could afford to lose. Snowflake is a promising young company with good long-term potential, which is really how you should be thinking about investing anyway.