Money Stuff: Slack Is Going Public, Whatever

BloombergOpinion Money Stuff

Matt Levine

Slack Today Slack Technologies Inc. will go public by a direct listing, with its stock opening for trading on the New York Stock Exchange. Normally a private company goes public by an initial public offering: It hires investment banks to market its stock to investors, the investment banks build a book of demand, and then, on the day of the IPO, the company sells a whole bunch of shares, all at once, all at the same price, to those investors. Often the company's pre-IPO shareholders—its early investors, employees, executives, etc.—will also sell some stock in the IPO, at the same price as the company. And then, for a while after the IPO, none of those early investors will sell any more stock: Usually the company and its pre-IPO shareholders will sign lock-up agreements promising not to sell any more stock for at least six months after the IPO. (The idea is that the bankers can tell the buyers in the IPO that no more supply will be coming for a while, that if you want stock you have to buy in the IPO.) And of course for a while before the IPO, none of them will usually sell, in part because the IPO is a big marketed event that represents the best chance to get a good price for the stock, but also in part because it is private stock and it's hard to sell; you can't just call up your broker and tell her to dump your stock. The IPO is a single salient point in time: There is no supply of stock for a while, then there is a big sale of stock all at once, and then there is no new supply again for a while. If you want to sell, or buy, the IPO is your best chance of doing that.  A direct listing does not work like this. The day of the direct listing is not, compared to the IPO, a particularly coordinated sale: Slack's bankers will have spent time calling potential buyers and gauging their interest in the stock, and they will have spent time calling potential sellers (early investors, etc.) and gauging their interest in selling, but they generally won't have firm orders on either side. Instead there will be an opening auction on the New York Stock Exchange, as there is for every NYSE stock every morning, and people will put in (or withdraw, or modify) buy and sell orders, and eventually a price will clear the market and the stock will start trading.  Equally important, there is no lock-up: If you are an early Slack investor and you want to sell your stock, you can put in an order to sell in that opening auction, or you can wait to see where the stock opens and sell 10 minutes later, or a day or a week or a month later. You can sell some of your stock now and some in an hour and some in a week; you can sell if the stock goes up and not if it goes down; basically anything that anyone can do with any stock, you can do with Slack. You can even sell before the direct listing! If you want? It is not super easy; Slack was not public or traded on an exchange, so you'd have to sell to accredited investors and go through some complexities to do it, but people could sell Slack stock last week, instead of this week, if they really wanted. In fact pre-IPO Slack investors sold about 7.1 million shares last month (after Slack announced its plans to go public), at an average price of $26.82 per share and a price range of $25.75 to $31.50 per share. That is not a ton of stock—I suspect Slack will trade more than that in its first 15 minutes as a public company—but it is $190 million worth of stock trading just a few weeks before the big liquidity event that is Slack's listing. (It was also substantially more than Slack traded in previous months.) Why sell stock last month when you could wait to sell it on the exchange this month? Well, why sell stock any month when you could wait another month? Maybe you think the price will go down, maybe you have bills to pay, maybe you want to diversify into another stock, whatever. The direct listing is not all that salient a liquidity event, so if you had some reason to sell last month, you just did. An IPO is a big step change in the trading of a stock: Before the IPO, it doesn't trade; after the IPO, it does. A direct listing is a particular point in a gradual process. Before Slack announced its direct listing, it traded a little bit. After that announcement, it traded a bit more. Today, it will trade some more. Tomorrow, maybe it will trade a bit more: Early investors will watch today's trading and, if they like what they see, they might sell more stock tomorrow. Or in a week. Eventually Slack will just look like a public company, with a shareholder base consisting of the sorts of large mutual funds that typically buy large chunks of shares in initial public offerings, instead of its current shareholder base of venture capitalists and early employees and so on. Today is an event in that transition, but it is not the event in that transition, and it might not even be the most important one. Maybe next week a couple of big mutual funds will decide it's time to buy Slack, and a couple of Slack's big venture capital investors will decide it's time to get out, and they'll quietly cross a few big trades that reshape the investor base and no one will notice. So the news is that Slack is going public today, but ideally that would barely be news: With IPO heavyweight advisers from Goldman Sachs Group Inc., Morgan Stanley and Allen & Co. helping to steer Slack through its listing alongside market maker Citadel Securities, all eyes will be on how the first day of trading plays out. But the company and its investors aren't looking for a meaningful stock pop -- and want to avoid the volatility -- that often accompanies high-profile share sales, according to a person familiar with the process. Slack isn't selling shares at all, and none of its pre-IPO investors are committed to sell at the opening price, so there's no particular harm if there's a big stock pop—no one is "leaving money on the table" or whatever—but, sure, if you have people who want to sell stock then volatility is bad. The reference price set yesterday was $26 per share, in line with trading last month; the first indicated opening range set by Slack's market maker this morning was $30 to $34 per share, a modest increase from private trading which makes sense given the greater liquidity and demand for public shares. (The indicated range has crept up since.) It's all just sort of boring, which is the point. Some fraud We talked the other day in sort of philosophical terms about cryptocurrency and money. One thing that I said is that the basic technology of modern money is the list. For most practical purposes, dollars are just a list of numbers maintained by banks; you have dollars to the extent that a bank has a number of dollars next to your name on the list that it keeps of dollars. This is a troubling thing to think about, generally; what if they lose the list? And so a lot of banking regulation is about deciding who can keep those lists and how they maintain them, and a lot of our norms and expectations about money are centered around understanding who can keep the lists and how we are supposed to trust them. Another thing that I said is that cryptocurrency offers a different way of keeping the list, one in which trust in intermediaries is replaced by a purely technological method of keeping the list and coming to consensus about who has the money. If you take this seriously, it might change your view of the traditional financial system's methods. If you get really into Bitcoin, you might stop trusting the traditional financial system to keep lists of who has money: It all just seems more arbitrary and opaque and hack-prone than the pure immutable code of Bitcoin. One possible result of this would be to stop trusting anyone to maintain lists of who has money—to keep all your wealth on the blockchain or in a sack of gold—but that is actually kind of hard to do in the current state of the world. Another, somewhat counterintuitive, possible result would be to just lose all perspective on whom to trust. You might conclude that all financial intermediation is arbitrary and end up taking your money out of a big traditional bank and investing it with, um, some random fraud. Here is a U.S. Commodity Futures Trading Commission enforcement action against an alleged "$147 Million Fraudulent Bitcoin Trading Scheme," and the alleged scheme is just as straightforward as can be: The defendants made numerous material misrepresentations and omissions, including that they (1) earned customers 1.5 % in daily Bitcoin trading profits and up to 45% per month; (2) used risk diversification methods to protect customers' Bitcoin deposits; and (3) provided a "safe haven" from Bitcoin market risks.  In reality, the defendants made no trades on customers' behalf, earned no trading profits for them, and misappropriated their Bitcoin deposits. The defendants concealed their fraud by providing customers with sham account balances and profit figures that falsely reflected trading profits that did not exist. The defendants allegedly "used the Control-Finance Website to generate automated profit credits that accrued in customers' accounts each day. While customers' Control-Finance accounts falsely reflected growing balances, in reality those accounts were empty." I love the word "automated" there. They went and built an algorithm that created automated trading profits by writing bigger and bigger numbers on a website. It is the simplest possible trading algorithm! It cuts out all the messy complicated trading and skips straight to the key step of writing down profits. It's good enough for your purposes, if your purposes are fraud. The other really fraud-friendly aspect of Bitcoin that you can see here is just the numbers. Not so long ago, if you wanted to run a fraud, you could offer people 45% monthly (!) returns, or you could offer them risk-free guaranteed returns, but if you offered them 45% monthly returns that were also free from market risk then some of the ever-so-slightly-more-sophisticated marks might say "wait this all sounds a bit too good to be true." Bernie Madoff ran a huge sustained Ponzi scheme by offering fake, steady, but reasonably modest returns. But Bitcoin has a fairly recent history of (1) fabulous gains and (2) wild volatility, so if you offer people fabulous gains and relative safety then it doesn't sound quite so crazy. "Wow 45% monthly gains with no risk is probably even better than the 60% monthly gains with total risk that I would otherwise expect," an investor might just about say, and before you know it you're raising $147 million based on nothing more than a website.  People are worried about bond market liquidity The basic worry about bond market liquidity is: Banks used to be bond dealers who took risks and used their balance sheet to intermediate bond trades. If you wanted to sell a bond, you'd call a bank and they'd buy it; if you wanted to buy a bond, you'd call a bank and they'd sell it to you. Finding the other side of the trade—re-selling the bond you sold them, or buying the bond they sold to you—was their problem, done at their own risk and on their own time, after they did the trade with you. You got a fixed price immediately. Now, with post-crisis regulation and risk aversion, banks are mostly just bond brokers who try to connect buyers and sellers. If you want to sell a bond, you call a bank and they put you on hold and call around looking for someone to buy it. If they find someone, they pick you back up, set up the trade, and take a cut of it; if they can't find anyone, then you don't get a trade. If it takes all day, it takes all day. You take the risk that they can't find a buyer, or that the price moves while they're looking for one. If you need to get out of a big bond position fast, that could be bad for you—for you, not for the bank, who is just a risk-free middleman. This story is not necessarily 100% true (banks were not totally selfless risk-takers before the crisis and are not totally risk-averse now), and to the extent that it is true it is not necessarily systemically bad (it probably is better for the risk of bond-price moves to be borne by mutual funds rather than banks), but it does sound straightforwardly inconvenient for bond investors. If you want to sell a bond, it is nice to just be able to call up a reliable professional who will immediately buy it at a competitive price; that is a valuable service, for you, that banks used to provide a lot of and now provide less of. But there is a completely different way of telling this story. Here is that version of the story: Some people—end users, real-money investors, mutual funds and pensions and insurers—want to buy bonds. Other people—again, real people—want to sell bonds. If they could just meet up with each other, then the sellers would sell bonds to the buyers, they would both be happy, and there would be gains from trade. But the market was not set up that way. It was hard for sellers to find buyers, and hard for buyers to find sellers. Instead, if you wanted to sell, you had to sell to a dealer; if you wanted to buy, you had to buy from a dealer; either way, the dealer got a big cut. And the dealer system stood in the way of the better, more efficient system of real investors trading directly with each other: The dealers controlled the market, they controlled the information and knew who the buyers and sellers were, they had the biggest inventory, they were the most willing to trade, and big clients had to rely on them and couldn't risk upsetting them by trying to cut them out of traders. There were all-to-all trading platforms connecting real buyers and sellers, but they were not all that useful for doing big trades, because big investors couldn't afford to break their dependence on the dealers. Then the dealers just stopped doing that job of intermediating trades And so now everyone can pursue the paradise of a first-best market structure in which real buyers and real sellers can find each other and trade directly, splitting the gains from trade themselves instead of paying a lot of money to dealers who control information.  I am not sure about this story! But here is a paper by Gideon Saar, Jian Sun, Ron Yang and Haoxiang Zhu titled "From Market Making to Matchmaking: Does Bank Regulation Harm Liquidity," which sort of makes the case for it[1]: Post-crisis bank regulations raised the market making costs of bank-affiliated dealers. We show that this can, somewhat surprisingly, improve the overall welfare of investors and reduce average transaction costs, despite the increased cost of immediacy. Bank dealers in OTC markets optimize between two parallel trading mechanisms: market making and matchmaking. Bank regulations that increase market making costs intensify competitive pressure from non-bank dealers and incentivize bank dealers to invest in technology that shifts their business towards matchmaking. Thus, post-crisis bank regulations have the (unintended) benefit of replacing the costly balance sheet of banks with a more efficient form of financial intermediation. The paper is mostly a theoretical model of bond-trading microstructure, rather than an empirical study. But the authors argue that it fits with the data. For instance, there are "empirical findings of an increase in the cost of immediacy together with a decline in average transaction costs around the enactment of post-crisis bank regulations": Investors pay more for the immediacy of using a bank dealer's balance sheet, but the pay less to trade bonds in general, since they are more likely to trade by matching up with each other and not paying a dealer to take risk. Similarly, there has been "an increase in matchmaking volume driven by the bank dealers," combined with "a significant decline in capital commitment to market making." And "overall trading volume in bonds has significantly increased as our model predicts," though that is ambiguous; a lot more bonds have been issued, so the increase in trading volume is not necessarily proof that trading has gotten easier. More generally, I think there are two ways to think about middlemen in markets. Sometimes you have a thing, and you want to sell it right now, and you want someone whose job it is to buy the things that you're selling who will immediately buy the thing at the market price. And when you don't get this, you complain. (About bond market liquidity.) Other times you look at the market as a whole in a period of leisurely reflection, and you count up how much money you spent buying and selling things over the past year, and you think, "man, for what? I paid all that money to middlemen to sell things, and other investors paid all that money to buy the same things, and if I'd just sold the things to the people who wanted to buy them then we'd have saved all that money." And you complain. (About rapacious high-frequency traders, for instance, in the equity markets.) The middlemen don't seem worth it, in those moments when you don't need them urgently. Things happen Deutsche Bank Faces Criminal Investigation for Potential Money-Laundering Lapses. Executive Turnover Clouds Deutsche Bank's Wall Street Future. Banks to Fight Over $18 Trillion Held By the Poorest of the Rich. Cryptocurrency Startups Are in Limbo as Regulators Grapple With Risks. The Future of Housing Rises in Phoenix. Fannie and Freddie's Latest Push: Factory-Built Homes. A British Fund Star's Fall Shows the Peril of Illiquid Holdings. What It's Actually Like to Be on House Hunters—Twice. Four Members Of One Texas Family Plead Guilty In Lucrative Masters Ticket Resale Scheme. Notorious B.I.G.'s 'one-room-shack' now costs $4K a month.  If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Or you can subscribe to Money Stuff and other great Bloomberg newsletters here. Thanks! [1] Technically they are making the case that regulation has caused a shift from market-making to matchmaking *by dealers*, and that this shift is efficient *for investors* because the dealers charge less for matchmaking than for market-making (because market-making is risky for them and matchmaking isn't). But the basic argument works just as well to explain a shift from market-making by dealers to matchmaking *by anyone*, including non-dealer all-to-all platforms that match up real investors directly without involving dealers as middlemen. Anyway here's an article about how "electronic bond-trading platforms are racing to expand their offerings as traders demand greater integration of the technology that is transforming how they buy and sell holdings," suggesting that that shift is real.

Bloomberg L.P. ● 731 Lexington Avenue, New York, NY 10022 Web ● Facebook ● Twitter Feedback ● Unsubscribe