Direct listings are an easy short.
I will build and present a case to delay long-only entry points into direct listings and initiate short positions based on activity in secondary markets and lockups.
Before we start, let’s dive into some basics.
Direct listings provide startups with an alternative way to go public than a traditional IPO or via a merger with a SPAC. In contrast to an IPO process where the company hires bankers, a direct listing allows the company to sell existing shares held by founders, employees, and early investors to public investors like us without issuing more shares.
The two main advantages of a direct listing:
- Fewer fees.
- Employees and insiders are not subject to a lockup period and can sell right away.
While these sound like great advantages to the company, what do they mean to us as investors?
Without JPMorgan, Goldman Sachs, and Credit Suisse, companies electing for a direct listing have the potential for better price discovery. We previously covered the topic of how investment bankers build a book for IPOs.
An IPO is effectively a Dutch auction; you get bids from dozens of investors and then price the IPO at the price that clears the market, meaning that the whole deal is priced off the lowest buyer’s price. — Matt Levine
A small group of institutional investors determines the price per share of an IPO, and they often underprice the shares resulting in a big pop on the first day of trading, dubbed the IPO pop. Many early-stage investors have written about the issue and how companies are leaving money on the table by electing to go through a traditional IPO.
Bill Gurley of Benchmark invested in UBER.
Alternatively, direct listings open up all of the company's shares to be traded, and the entire market, including us, determines the price. In either scenario, we have stock of a private company that has been thinly traded all of a sudden available to everyone. The dynamics of supply and demand that exacerbated the risks of buying an IPO is still present with a direct listing.
In an IPO, the initial offer is exclusive to institutional investors who have access to the investment banks, and there’s often an immediate benefit to those relationships. We cannot buy the shares until they start trading on the secondary markets. Nor can insiders, and most early investors sell their shares until a lockup period is over. These components often add up to create an imbalance where demand greatly exceeds the supply for the company's shares.
Recently, Warren Buffet bought $250 million worth of Snowflake stock at the offered IPO price of $120 per share. On the first day of trading, SNOW hit highs of $319 per share, resulting in paper gains of $1 billion for Buffet for less than a few months into their investment. 😤
The reversal of supply and demand
In a direct listing, no secret handshake is needed to get into the offering. Simultaneously, all the employees and early investors who have stood behind the company are eager to reap the rewards. Without a lock-up provision or inclusivity, there can easily be more people willing to sell their shares than available buyers.
A case for lock-ups
Can’t companies implement lock-ups for direct listings? Two companies have chosen to go this route out of the four high profiled direct listings so far. Slack and Asana had no lock-up in place, while Spotify restricted the sale of stock from Tencent, which owned 7.5% of Spotify’s common shares at the direct listing time. Palantir took an even more austere stance and decided to limit the sale of approximately 76% of their stock—only 383,609,648 shares were allowed to be traded out of a total of 1,625,305,107 shares outstanding.
Palantir’s Amended S-1
What happens if we look at the first month of trading for each of the four companies? Will we see an effect the different lock-up policies had?
I pulled the historical data from Yahoo! Finance and calculated the compounded monthly growth rate for each stock (Here is the spreadsheet). I had to extrapolate PLTR and ASAN since both stocks have not yet traded for a month. I decided to look at the first month because the time period is short enough that there’s usually no material disclosure or changes to the business, such as an earnings report that may introduce company-specific bias.
Note: the extrapolations are usually skewed towards the negative side, and I would expect that ASAN won’t end up quite as far in the negative, and PLTR would probably see more positive growth.
Slack and Asana, the two companies without lock-ups, show a negative growth rate or a drop in the stock’s value for the first month. Spotify and Palantir with the lock-ups, show an appreciation in stock price. While data is minimal, this seems to hint at our hypothesis so far. But is a 7.5% lockup really enough to result in Spotify having a big positive growth rate?
How much is really liquid?
Outside looking in, if a lock-up provision is not in place, all of the company's outstanding stock can be transacted. However, this only works in theory but not in practice. Executives and members of leadership at the company need to be incentivized to stick around. If the CEO can immediately sell all of her shares, what would align her interests in the long term with the company? For this very reason, many companies have ownership guidelines in place for their executives. While I haven’t been able to find any public information on existing guidelines for any of the four companies, I don’t think it’s unreasonable to assume that at least half of the executive team shares have some internal restrictions. Spotify’s lock-up might actually be larger than 7.5% in practice, and both Slack and Asana may not be completely freely traded. This leads me to think that there is less of a binary relationship between a stock’s price movement and a lock-up policy in place. Instead, I think there might be another factor in play.
Activity in the private markets
Going public isn’t the only way startups can sell or transact their shares. A company can run a tender event where current investors and shareholders of a private company can sell their shares in a private secondary market. Those markets' advantage is to provide early liquidity and a method to de-risk for employees and venture capitalists. The disadvantages is that market participation and liquidity is low and can result in mispricing.
Why do secondary markets matter?
Secondary markets provide immediate liquidity to insiders. The insiders could be employees buying new cars or putting a downpayment on a home purchase. They could also be early investors who need to return capital to their limited partners. New investors offer the existing insiders to buy their shares at a negotiated price point. Having secondary market transactions before the company is public reduces the immediate sense of urgency to sell on the first day of trading since most of the investors buying into the secondaries are planning to invest long-term and do not need to sell at the direct listing price.
Let’s take a look at secondary transactions for the same four companies above.
Spotify transacted a total of 13,130,089 shares in a period of 9 months before the direct listing representing about 7.4% of the total 176,976,280 shares outstanding. If we break down the transaction ranges, we arrive at a weighted average share price of $94.21 on the secondary markets. The company has been trading during the whole time; the public price had never dropped below this weighted average price.
Spotify private transactions S-1
Slack transacted a total of 10,449,706 shares in a period of 8 months before the direct listing representing about 2.08% of the total 502,136,716 shares outstanding. The transactions had a weighted average of $16.86 per share and again while the stock never publicly traded below this price.
Slack private transactions S-1
Palantir ran secondaries earlier and had a total of 20 months of reported transactions before the direct listing. During that time, 60,893,422 shares, or approximately 3.75% of the 1.625 billion shares outstanding, changed hands at a weighted average price of $5.14 per share.
Palantir private transactions S-1
Asana had the least amount of secondary transactions. Over what is presumably 6 months, only 713,715 shares were sold on the secondary market, making up less than 1% of the 151,315,622 shares outstanding.
Asana private transactions S-1
Correlation or causation?
If we compare the % of the float that was transacted on the secondary market side by side to the first month returns, we see a distinguishable pattern.
With the least amount traded in the secondaries, Asana performs the worst, followed by Slack, Palantir, and Spotify. At the same time, if we remember that Palantir and Spotify had lock-up periods, they were also trading up during the first month. What if we look at the volume of shares transacted on the first day of trading? If our theory around the secondary market holds true, we should see a relationship between secondary transactions compared to the volume traded.
Asana vs Palantir first-day trading volume from Nasdaq
Spotify vs Slack’s first-day trading volume from Nasdaq
The volume traded for each company on the first day was divided by the total shares outstanding to arrive at a % of float traded on the first day.
Spotify with the largest % of its float sold privately had also the lowest % of the float traded on the first day. Palantir with the second-largest % of its float sold private had the second-lowest %traded on the first day. Asana and Slack bring up the bottom with roughly an equivalent % of their float being traded.
At this point, it’s still too early in the trading period for Asana and Palantir, and the sample size of four companies is too small to decisively conclude that we’ve uncovered a way to predict the price action of direct listings. However, it’s hard to shake that the quantitative story is matching up with the qualitative. Direct listings are still very new and it’s not surprising that it will take a few iterations to get right or at least to a state where it is actually better than a traditional IPO. I am using this initial groundwork to develop an investment thesis and a strategy for investing in direct listings in the future. Stay tuned for that future post but in the meantime, here are some takeaways for today:
Trade idea 1: delay longs
Great, we like a company that’s planning a direct listing. Read through the S-1 and take a look at lock-up provisions and private transactions. If there’s no lock-up provision and only a small amount has been transacted publicly, wait 1–2 months on the sidelines before initiating a position. When we start, we should dollar cost average in.
Trade idea 2: Puts on week 2
I personally do not think Asana is differentiating enough as a company and as a product for the company to merit a $5B market. Fundamentals aside, buying put options on week 2 after a direct listing is effectively the only way for us to short the stock—typically, it takes a week for the contracts to be available for purchase on Robinhood after the first day of trading. When buying a put, choosing a shorter duration that expires before the next earnings report will limit the risk that strong fundamental news will cause the stock to swing favorably. For this strategy to work, it’s too risky for out-of-the-money puts even though they are cheap. Instead, buy at-the-money puts.