Header Ads

Money Stuff: Green Shoes Look Funny

Money Stuff
Bloomberg

Greenshoes

Here is a fun weird paper (and related blog post) about greenshoes, by Patrick Corrigan of Notre Dame. It is titled "Footloose with Green Shoes: Can Underwriters Profit from IPO Underpricing?" We talk about greenshoes from time to time. Here is the way greenshoes are supposed to work. A company does an initial public offering, say of 10 million shares at $40 each. It will give its underwriters an option, called an "overallotment option" or more commonly a "greenshoe,"[1] to buy an extra 1.5 million shares (15% of the deal). On the evening that the IPO prices, say it's a Tuesday, the underwriters will allocate 11.5 million shares to investors, and will buy 10 million shares from the company. The underwriters will be short 1.5 million shares, the shares underlying the greenshoe: They have sold more shares, in the IPO, than they have gotten (so far) from the company. 

The next day, Wednesday, the stock will open for trading. Perhaps it will open at $42. Perhaps then it will start dropping. It will reach $40. There is a risk that it will go down further, "breaking" the IPO price of $40. The underwriters will step in. They will start buying stock. They will buy back stock at $40 to stabilize the stock at or near the IPO price. Hopefully the stock will start going up again and everyone—the company that did the IPO, the investors that bought it, the banks—will be happy. If not, though, the banks will keep buying until they have bought 1.5 million shares, the amount of the greenshoe. In that case, the banks have sold 11.5 million shares at $40 on Tuesday night, gotten 10 million shares from the company at $40 on Tuesday night,[2] and bought back 1.5 million shares in the open market on Wednesday. They are net flat; they own zero shares and have made zero dollars of trading profits.

This is a legal form of market manipulation. It is generally believed, in U.S. equity capital markets, that you need to have a greenshoe: The banks need to have "ammunition" to defend the IPO price; investors who buy in the IPO want that sort of guarantee that, if the stock starts dropping, there'll be someone there to buy. This is called "stabilization," and there are some rules about how it works—roughly, the banks are allowed to keep the price from falling, but they're not supposed to push it up—but it is allowed. It's a little weird, but allowed.

Most IPOs go up, though, not down. In that case the banks do not have to stabilize, and they do not buy their 1.5 million shares back in the market. If the stock opens at $42 on Wednesday and quickly climbs to $60, the banks seem to have a big loss on their short position. This is no problem, however. They have that greenshoe option. They wait a day or two to make sure the stock really isn't going down, then call up the company and exercise the option; they buy another 1.5 million shares from the company at $40.[3] They sold 11.5 million shares at $40 on Tuesday night, got 10 million shares at $40 from the company on Tuesday night, and got 1.5 million more shares at $40 from the company on Friday morning. Again, they are net flat.

So that's the theory. The greenshoe is an option that the company gives the banks, but it's not for the banks. The banks don't pay for the option, and they don't make a profit off of it; the option is just a technicality that allows them to stabilize the stock without taking big trading risks themselves. The greenshoe is a reassurance for investors in the IPO, not a windfall for the banks.

Ha, well, mostly. This can go wrong. The way that it has notably gone wrong in recent years is: What if the stock opens, not at $42, but at $30? If the opening trade in the stock is well below the IPO price, the banks can't really "stabilize" the stock by buying at the IPO price. They will do their best to stabilize it, but at the new, lower price. They'll buy back the 1.5 million shares at, say, $30. They sold 1.5 million shares short at $40 and bought them back at $30, making a trading profit of $10 a share. 

This happens from time to time; it is embarrassing for the banks but also, obviously, nice to make some money. High-profile cases include the IPOs of Facebook Inc., Lyft Inc. and Uber Inc.; in all of them, there was some complaining that the banks made a windfall by overpricing the IPO and then profiting on their short shares.[4]

But of course most IPOs go up. These days, IPOs frequently double on the first day, often on the first trade. There is in theory another way for the greenshoe to be a windfall for the banks when the stock goes up. It's embarrassingly simple. It's just: Look, the greenshoe is a call option to buy a bunch of stock at the IPO price. The IPO price is probably too low. That's a valuable option. What you do is, on Tuesday night, you allocate 10 million shares to investors at $40, instead of 11.5 million. You get 10 million shares from the company at $40. You also get a 1.5 million-share greenshoe option, but that's your business. The next day, the stock opens at like $80. You shrug "guess we don't need to stabilize," you exercise the greenshoe option, and you get 1.5 million more shares for $40. Then you sell them, for $80.[5] You've made an extra $60 million. You priced the IPO too low, you got an option to buy shares at that low price, the stock went up, you exercised the option, you made a bucket of money.

There is a synthesis of these approaches, which is: On Tuesday night you allocate 10.5 million shares at $40 (and get 10 million from the company, leaving you short 0.5 million).[6] Then on Wednesday, if the stock goes down, you buy 0.5 million shares for $30 or whatever, making $10 per share on your 0.5-million-share short position, and you do not exercise the greenshoe. If instead the stock goes up, you exercise the greenshoe, getting 1.5 million more shares at $40; you deliver 0.5 million of those to close out your short and sell the other 1 million for $80 or whatever. Either way you make money. The more the stock goes up or down, the more money you make. The greenshoe is a call option, and call options are more valuable the more volatile the stock is. A call option in an IPO that will either double or fall by half is very valuable, particularly if you get it for free.

Now, those last two paragraphs are crazy! Everything I said before them was kind of normal, just a description of how greenshoes work, but those last two paragraphs are very much not how greenshoes work. The greenshoe is not just a free valuable option for banks to trade and make money with. You don't allocate 10 million, or 10.5 million, shares of a 10-million-share IPO and then hope to make a profit by exercising the greenshoe and selling stock at the higher post-IPO price. I used to be an equity capital markets banker and that was very much not how we did it. I informally polled a few ECM bankers and lawyers about whether they'd ever heard of such a thing, and their answers ranged from "lol no" to "lol no that's illegal."[7]

Anyway Corrigan's argument is that it's not actually illegal: There are a lot of rules that seem like they would prohibit banks from doing that, but he argues that they don't quite apply, and a bank really could just trade the greenshoe for its own profit.[8] He also suggests that they do, and that this is why IPOs go up and down so much:[9]

Underwriters have incentives to operate the bookbuilding process and the IPO pricing negotiation with an aim of underpricing or overpricing IPOs, but not pricing them accurately.

The prediction is precisely the opposite of the price stabilization theory: IPO initial returns should exhibit variability, not stability. Consistent with this prediction, we observe that IPOs with negative initial returns have mean first day returns of 8.1 percent, while IPOs with positive initial returns have mean first day returns of 25.5 percent.

Thus, the blowouts in the Airbnb IPO and the flop in the Wish IPO are two sides of the same coin. In the former, underwriters monetize a very valuable call option. In the latter, underwriters monetize a valuable short position.

Again, I don't buy that this is what banks actually do, or that when, like, Affirm Holdings Inc.'s IPO doubled on its first day of trading its banks made an easy $178 million windfall.[10] I think that banks sold all their greenshoe shares at the IPO price on the evening of the IPO, and then made no money on the greenshoe. On the other hand, when IPOs do flop right out of the gates, banks do—shamefacedly and reluctantly—make some money.

Corrigan's conclusion is basically that it should be more illegal for banks to do this sort of thing. And fine, I guess; I don't think that they do this sort of thing, because they already think it's illegal, so making it more illegal wouldn't hurt much.[11] But he also argues that the existence of the greenshoe makes IPOs more volatile: IPO underwriters make money when IPOs are dramatically overpriced (true), and when they are dramatically underpriced (probably not true).

Meanwhile the point of the greenshoe is to stabilize post-IPO trading prices and, uh, that does not seem like all that compelling an explanation? Most IPOs go up, so the underwriters don't have to stabilize them. Of the ones that go down, most go down pretty quickly, so it's not clear how much good the stabilizing does. Corrigan:

Contrary to the price stabilization theory's predictions, a surprisingly high number of IPOs break issue very rapidly. More than 23 (28) percent of IPOs in my sample traded below the initial offering price by the end of the first (tenth) day, with an average return of negative 8.1 (negative 10.7) percent. More than 16 percent of IPOs traded below the initial offering price in the very first trade on exchange, with an average decline of negative 6.2 percent.

That is, a majority of IPOs that fall below the IPO price do it on the very first trade, before the underwriters even have a chance to stabilize. Also the theory behind stabilizing is a little weird. Corrigan notes that "the theorized warranty mechanism takes the form of a wash trade and does not inject any new demand into the aftermarket": Instead of selling 11.5 million shares in the IPO and then buying back 1.5 million the next day to stabilize the stock, the underwriters could just sell 10 million shares and let whoever was going to buy those other 1.5 million shares buy them the next day.

The broader point may be that greenshoes just aren't that necessary anymore, that they are vestiges of the olden days in the capital markets. Companies find it annoying to have uncertainty about how many shares they are selling, and to give banks this weird option. With the rise of alternatives to the traditional IPO, it is harder for banks to argue that the greenshoe is necessary; with all the huge IPO pops these days, it is harder for investors to argue that they need the protection of underwriter stabilization. You can just do an IPO without a greenshoe. DoorDash Inc. did. It was fine. If you are doing an IPO and you don't like what your banks are asking for, you don't have to do it. The power is with the companies now.

Proof of life

One dumb category of things that we sometimes discuss around here is, like, "things that are material to stock prices, yet weird." Like there is a simple model of stocks where what matters to a company's stock price are (1) its quarterly earnings and (2) announcements of major corporate events like mergers. Everyone knows these things are material, companies announce them in official press releases and securities filings, and they take pains to keep them secret before they are announced. If you know those things in advance, you can profitably trade the stock: If you know earnings will be good or the company will be acquired, you can buy stock (or short-dated out-of-the-money call options) before the announcement, and the stock will reliably go up after the announcement. But since you're probably not supposed to know these things in advance, you are probably doing illegal insider trading.

But in the actual world stocks move around a lot for reasons that are dumb yet predictable. If you were hanging out with Elon Musk last Thursday, and you watched him type a tweet saying "Use Signal," you could have said "hey hang on a minute before sending that," and then bought a bunch of stock of Signal Advance Inc. Signal Advance does not make Signal, the messaging app that Musk was presumably tweeting about, but his tweet definitely did drive Signal Advance's stock up 5,100%. That was pretty predictable; if Elon Musk tweets something that sounds like a company, that company's stock will go up. And so people sometimes ask me questions like: If you knew about that tweet in advance, could you trade the stock of Signal Advance? Could Elon Musk himself have traded it? 

I don't know? I kind of feel like the answer is "sure whatever go ahead," but that is not legal advice and I do not feel strongly. The point that I sometimes make is that, in U.S. securities law, there is a standard of materiality, in which a fact is material if it would be significant to "the reasonable investor"; and then there are actual things that make stocks go up and down, which are not always, you know, reasonably significant. Elon Musk's tweet was unquestionably material to Signal Advance's stock price, but it surely would not have been significant to a reasonable investor. It had nothing to do with Signal Advance's long-term cash flows or whatever, and anyway within hours of the tweet it was quite clear and public that Musk wasn't talking about Signal Advance. The stock kept going up anyway, for days, because markets are not always reasonable.

Anyway if you were like Jack Ma's dry cleaner, and he came by to pick up some pants this weekend, you had billions of dollars' worth of market-moving information:

He appeared for less than a minute and said nothing about the Chinese government clampdown that had left his business empire in crisis.

But for investors who'd been waiting months to catch a glimpse of Jack Ma, the entrepreneur's participation in a live-streamed video conference on Wednesday was enough to trigger a $58 billion sigh of relief. That's how much Alibaba Group Holding Ltd.'s market value soared after a clip of Ma speaking to a group of teachers began circulating online -- his first public comments since disappearing from view late last year.

Much about the future of China's most famous businessman remains unclear. Yet analysts said Wednesday's video was a sign that worst-case scenarios -- such as jail time for Ma or a government takeover of his companies –- are probably now off the table. It's unlikely Ma would have participated in the event without at least tacit approval from Beijing; state-run media including the Global Times were among outlets that posted snippets of his talk or wrote stories about his appearance. …

"Alibaba is not out of the doghouse, but at least it's clear that the current anti-monopoly drive is not about punishing Jack Ma," said Zhang Fushen, senior analyst at Shanghai PD Fortune Asset Management.

Speculation about Ma's whereabouts had intensified in recent weeks after it emerged that he skipped the recent taping of a Shark Tank-like TV program that he had created. Chinese authorities have in the past quietly detained billionaires that run afoul of the Communist Party.

To be fair, it is probably material to a reasonable investor that Alibaba's founder is alive and not imprisoned somewhere! Still it … requires explaining. It is not in the textbooks. An investor focused on cash flows and corporate events might not have been paying attention to Ma's television appearances, and might not have noticed that he was gone. That investor might not have found a one-minute video about teachers to be all that relevant to a $700 billion company. But the market did.

Insider trading

Or, if you know a guy is going to go on TV and talk up a stock, is it illegal to trade that stock? Sure, I guess, I don't know, a little, whatever. It depends. Like if you are a hedge fund manager, and you know you are going to go on TV tomorrow to talk about stocks, you can probably buy the stocks that you plan to talk about. Ideally you would disclose, on the TV show, that you own the stocks. But when you buy the stocks you are trading on your own proprietary information, information about your intentions; it can't really be illegal to do that. 

But if you work for the TV show? If you're a host or reporter for the show? Then … well, I guess it will depend on your arrangement with the show, but in many cases the show will take a dim view of you buying short-dated call options on the stocks you are going to recommend the next day. Your recommendations are in a sense the property of your employer; you have some vague fiduciary relationship with your viewers; it would be somehow gross for you to front-run them. And in U.S. securities law, in fact, there is a famous case in which a Wall Street Journal columnist named R. Foster Winans was convicted of insider trading for leaking the contents of his column—which picked stocks—to a stockbroker, who traded on them before they were published. Those columns, which Winans wrote, nonetheless belonged to the Journal, so he wasn't allowed to trade on them, or let other people trade on them.

Meanwhile in India:

The Securities and Exchange Board of India (Sebi) on [last] Wednesday barred a television (TV) anchor Hemant Ghai, his wife and mother from accessing the capital markets for indulging in fraudulent trading activity.

The market regulator has alleged that the three individuals pocketed nearly Rs 3 crore between January 2019 and May 2020 by dealing in stocks that were being recommended on TV show Stock 20:20 on CNBC Awaaz, a leading business news channel.

"It was observed that Jaya Hemant Ghai (wife) and Shyam Mohini Ghai (mother) have undertaken a large number of Buy-Today-Sell-Tomorrow (BTST) trades during the relevant period in synchronization with the recommendations made in the Show. Shares were bought on the previous day to the recommendations being made on the stock 20-20 show and sold immediately on the recommendation day," said Madhabi Puri Buch, whole time member (WTM), Sebi in an order.

Here is the SEBI order. Hemant Ghai was the co-host of the show, "which airs on trading days at 7:20 am and it recommends certain stocks to be bought  during the  day." The day before it aired, his wife and mother would buy shares in the stocks that were going to be recommended. You can, and SEBI did, draw some obvious connections. Not allowed!

Things happen

Morgan Stanley Traders Drive Earnings Surge for Best Year. DOJ Inquiry Into Sen. Richard Burr's Stock Trades Ends Without Charges. Trump Pardons Levandowski, Who Stole Trade Secrets From Google. Office Depot Rebuffs Takeover Offer From Staples. US finance chiefs weigh how to spend vast corporate cash piles. Chinese Investors Pull More Than $3 Billion From Funds That Missed Out on Ant IPO. The Dark Side of Investor Conferences. The Dunning-Kruger Effect Is Probably Not Real. "In the pandemic's early days, photographer Thomas Smith funneled heat through a tube from a computer he uses mainly for mining bitcoin to maintain a small greenhouse in his garage."

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] Named after the first company to do it, Green Shoe Manufacturing Co., in 1963.

[2] Really it's more like $38: The banks' fee for doing the IPO is expressed as an "underwriting discount," so they buy shares from the company at $38 or whatever (typical IPO fees range from 1 to 7%) and sell them for $40. I am ignoring the IPO fees in the text but of course they exist.

[3] Again, really $38 or whatever; the banks get the underwriting discount on the greenshoe shares.

[4] In fact it's a bit worse than this. Underwriters are allowed to go "naked short" in an IPO. In our IPO example, instead of laying out 11.5 million shares (10 million from the base IPO and 1.5 million from the greenshoe), the underwriters might lay out 12.5 million shares in the IPO. Then they are "naked short" 1 million shares. If the stock threatens to go down, they buy back those extra 1 million shares at or below the IPO price, giving them more "ammunition" to stabilize the price. If the stock goes up, they just lose money—their own money—on the naked short. Banks don't do this very often, and if they do, it's because they worry the stock might go down. Then, of course, if it does, they make more money, and it's more embarrassing. This happened in the Uber IPO.

[5] Or you sell them for $80 first—as short sales—and then exercise the greenshoe to buy them back at $40.

[6] I don't know the right proportions here. You could try to do some Black-Scholes math to figure out the theoretical delta of the greenshoe option, and then sell that many shares. That does not seem especially like the right approach. What I might do is say, look, about 80% of IPOs go up and about 20% go down, so I want to be short about 20% of the shares underlying the greenshoe. That would argue for being short about 0.3 million shares. I just used 0.5 million—one-third of the shares—in the text because it's an easy-ish number.

[7] To be fair, sometimes banks will lay out less than 115% of the deal if it's very hard to find buyers, but then they'll cancel a portion of the greenshoe rather than hope to trade it for a profit. (Also in that case you will usually not see a big pop, since no one wanted to buy the IPO.)

[8] One important rule here is what is called the "free riding and withholding" rule, which prohibits banks from keeping IPO shares for themselves to sell at a higher price. If you do a 10-million-share IPO, with a 1.5-million-share greenshoe, and you expect the IPO to pop, you should actually allocate, like, *seven* million shares on Tuesday evening. Then on Wednesday the stock doubles and you sell the extra 3 million shares (plus the greenshoe) for a huge profit. But *that* is clearly illegal: If you underwrite an IPO, you have to sell all the shares at the IPO price; you can't keep them back to try to sell later after the IPO pops. One view is that this rule also applies to greenshoe shares; Corrigan claims it does not.

[9] I don't entirely understand his mechanical description of how he thinks underwriters make money—he says they have "a straddle position," which strikes me as not at all right—but he suggests roughly that they short lots of shares when they know the IPO will go down and short no greenshoe shares when they know it will go up. This requires them to know the future. My description in the text, in which the underwriters short some but not all of the shares underlying the greenshoe, seems like a simpler and more theoretically correct way to make money on IPO volatility using the greenshoe. 

[10] Affirm did a $1.4 billion IPO with a greenshoe of 3,690,000 shares; the IPO price was $49 and the stock closed at $97.24 on the first day. So (97.24 - 49) x 3,690,000 = about $178 million.

[11] Also many underwriting agreements make it clear that the greenshoe is "solely to cover overallotments," and Corrigan concedes banks aren't allowed to treat that sort of greenshoe as a free option. He just argues that agreements that don't say that can be treated as a free option. One solution is just to make sure the underwriting agreement says that.

No comments