Yet more Uber
“Uber Blame Game Focuses on Morgan Stanley After Shares Drop,” is the headline here, and obviously Morgan Stanley doesn’t look great after pricing Uber Technologies Inc.’s initial public offering near the low end of its marketing range and then watching the stock fall 18% in its first two days as a public company. But I want to defend Morgan Stanley a little bit. Once a company is public, its stock price is determined by the market, and no bank has that much ability to control the price. Right now it looks like the market has decided that Uber’s stock price is $37 and change, for a market capitalization of a bit more than $60 billion. That is a disappointing number for a lot of people, 1 and you can debate whom to blame for that, 2 but you can’t put most of the blame on Morgan Stanley. Uber is worth what it’s worth.
Morgan Stanley’s error was marketing and pricing the IPO for more than Uber is worth, and that is a serious and embarrassing error that will no doubt leave many of its investor clients angry, but honestly what else could they have done? If Morgan Stanley had had perfect foresight and told Uber “we are going to market this deal at a price range of $33 to $37, price it at $35, and see a modest pop as it trades up to $37 in the next few days,” they would have been laughed out of the room, and someone more optimistic would have led the IPO. (And Morgan Stanley would not have collected $40 million in fees. 3 ) A $35 IPO would give Uber a (post-money) valuation of $59 billion, below its value in 2015, and less than half of what Uber’s banks actually pitched last year. (It also would have raised $1.8 billion less for Uber, which seems important given that Uber keeps losing money.) It would have been basically impossible for Morgan Stanley to convince Uber to launch the most anticipated U.S. technology IPO in years at such a defeatist valuation, and anyway it might have been self-fulfilling: If you launch the Uber IPO on the premise “the unicorn bubble is over, sorry everyone,” then who will want to buy?
Morgan Stanley’s job, as an IPO underwriter, was to intermediate between the desires of the issuer and its private investors, on one hand, and its new public investors, on the other hand. Half the job is to manage the expectations of the company and its private investors, getting them to give the public market what it wants in terms of price and structure; the other half of the job is to stir up demand from public investors, getting them to give the company what it wants in terms of IPO price and aftermarket trading. The problem with Uber is just that the public and private views of Uber were very far apart, and there’s only so much a bank can do to bridge them. Uber didn’t want to sell for $60 billion, the public didn’t want to buy at $100 billion, and so Morgan Stanley was more or less bound to look bad.
I mean, don’t get me wrong, you can quibble with some of Morgan Stanley’s choices, and people do. Here’s a quibble from “one investor at a multibillion-dollar shop”:
He said he grew suspicious days before the pricing because the syndicate of banks kept seeking reassurances that his firm wouldn’t flip the stock. Yet, the bankers also kept telegraphing there were ample retail investors hoping to buy in after the debut, which could cause the price to “pop” at least briefly, offering a chance for a quick and easy profit, the investor said. His firm ended up slashing its final order.
To be fair, that is every IPO: The banks want the stock to pop on the first day, but they also want most of the buyers in the IPO to hold on to their stock and not take that quick profit. If there’s enough enthusiastic long-term demand, it all works out. If there isn’t, it doesn’t.
Or here’s another quibble:
At least one of Uber’s largest investors, now in the red and speaking under the condition of anonymity, voiced frustration, suggesting the bank should have propped up the price more from the start. Yet that could have left the investment bank with less firepower to support the stock if it were to keep sliding in the days that followed.
That’s actually a tough one. Uber sold 180 million shares to the banks on Thursday evening, but the banks sold 207 million shares to investors in the IPO. The banks were short the extra 27 million shares (at $45 each). If all went well, they’d exercise their “greenshoe option” and buy the extra 27 million shares from Uber’s early investors at $44.41 each (the IPO price less the bankers’ fees), making an extra $15.9 million in fees for the banks and giving people like Travis Kalanick and SoftBank Vision Fund $1.2 billion of extra IPO proceeds. If things went poorly, Morgan Stanley—as “stabilization agent” on the deal—would instead buy back those 27 million shares in the open market, at or below $45, in an effort to support the stock price.
The problem is that Morgan Stanley never got a chance to defend the $45 deal price, because the stock opened at $42 on Friday (its first day of trading) and never got back to $45. It is unclear exactly how Morgan Stanley has done its stabilizing, but there were no great options. The stock opened at $42 on Friday morning and climbed above $44 for a while in the early afternoon; Morgan Stanley could conceivably have bought back all 27 million shares then to try to maintain that price, but then it would indeed have “less firepower to support the stock if it were to keep sliding,” as it did. Whatever support Morgan Stanley provided on Friday afternoon did not stem the decline. On the other hand, if Morgan Stanley did keep back any firepower, it didn’t do much good; the stock opened below $40 on Monday and has never gotten back there.
Awkwardly, if an IPO opens this poorly, the banks make more money stabilizing it. If Morgan Stanley did buy back all 27 million shares early Friday afternoon, when the stock was within shouting distance of the IPO price, it probably made $25 or $30 million of trading profits for the syndicate, about twice what it would have made by exercising the greenshoe if the stock had traded well. 4 If it held off, though, it made even more. If Morgan Stanley did half its stabilizing on Friday afternoon and the other half Monday morning, then the underwriters would have made about $107 million in trading profits on the greenshoe, 5 or a bit more than they made in IPO fees. It does not make up for the embarrassment and ill-will of an IPO like this, but it is some consolation.
Oh meanwhile this is insane:
[Uber] also could try to convince an existing shareholder such as Saudi-backed SoftBank, which has already invested $8 billion, to buy more shares at the depressed price. At a minimum, SoftBank may feel pressure to ask the companies in its investment portfolio that compete with Uber to ramp down cash burn, said one major Uber shareholder.
“Should index funds be illegal,” I often ask, on the theory that investors who own all of the competitors in an industry will encourage those firms not to compete against each other too strenuously. A common objection to this theory is that actual mutual funds don’t do that, particularly not index funds; they don’t call up companies and say things like “hey stop spending money and raise your prices.” But the SoftBank Vision Fund is among (many) other things sort of an index fund of ride-sharing companies, and its practices seem … different … from those of other big investors.
Are leveraged loans securities?
If you go to a bank and ask for a loan, the bank will ask you some questions to decide whether you’re likely to pay back the loan. You should answer those questions honestly; if you lie, that’s fraud. But coming up with the questions is the bank’s problem. If you just lost your job and want a $100,000 loan to go play roulette in Vegas, and you go into the bank and say “can you lend me $100,000,” and they forget to ask you questions like “do you have a job” and “what are you going to do with the money,” then you don’t have to volunteer the answers. These are not really issues that come up much in daily life. Banks want to be paid back, and they are repeat players with a lot of experience in evaluating loans, so they tend to ask the right questions. There are exceptions. 6
On the other hand, if you decide to raise money by selling stock in your company to the public markets, you have higher obligations. You have to write a big long disclosure document and send it to investors, and there are lots of rules about what it needs to include—audited financial statements, a detailed description of your business and strategy, disclosure about your executives and directors, what you’re going to do with the money—and if it turns out that that document “contained an untrue statement of a material fact or omitted to state a material fact required to be stated therein or necessary to make the statements therein not misleading” then you get in trouble and the investors can get their money back. The burden is on you to tell investors everything they need to know; it’s not on the investors to figure out what they need to know and then ask about it.
These two very different paradigms have an obvious explanation. The archetypal bank lender is a bank. The bank is bigger than you are, and knows more about money than you do, and has a lot more money than you do. The archetypal securities purchaser is an amateur retail investor investing his modest retirement savings. If you are selling stock in your company, he is smaller than you are, and knows less about money than you do, and has less money than you do, and can’t necessarily see through your tricks. You have to look out for him, but the bank can look out for itself.
But those are just the archetypes. In fact most of the money in securities markets, these days, is managed by full-time professionals, but they still get to rely on a disclosure regime that was originally designed for small retail investors. 7 This is not a bad thing; transparency and uniform information seems to make those professionals’ jobs easier, build confidence in the system, encourage deeper and more liquid securities markets, and create, if not a level playing field, at least a high baseline for everyone.
On the other hand, a lot of the money in the market for bank loans to companies is managed by those same professionals. Bank loans, once, were made by banks, which held them to maturity and had deep relationships with the companies who borrowed from them. Now bank loans are syndicated by banks, which go out and sell them to a group of investors. Those investors include lots of banks, but they also include hedge funds and collateralized loan obligations and other non-banks. Some of them will hold to maturity, but there is an active trading market for syndicated corporate loans. Bank loans can feel a lot like secured, floating-rate bonds, and in fact many credit investors buy and trade both bonds and loans.
“This is not necessarily obvious to everyone,” I wrote not too long ago, “and the historical difference between bonds and loans—bonds were sold by investment banks to investors, loans were made by commercial banks on their own balance sheets—still motivates some of the discussion and regulation of lending.”
For instance: Loans are not securities. If a company wants a loan, it doesn’t hire underwriters and prepare a prospectus. It hires arrangers and prepares an information statement, sure, but those are not the same thing. There’s less information in the information statement than there would be in a prospectus. The arrangers have less responsibility than underwriters do; they don’t vouch for the company to other lenders the way underwriters vouch to investors. 8 And the lenders—both banks and hedge funds—are charged with doing their own due diligence. 9
Here’s a story titled “Investors Suing JPMorgan May Redefine the Leveraged Loan Market”:
A group suing JPMorgan Chase & Co. and other Wall Street banks over a loan that went sour four years ago is alleging the underwriters engaged in securities fraud. If successful, the lawsuit could radically transform the $1.2 trillion leveraged lending market.
The defendants say there’s one key problem — unlike bonds, loans aren’t securities. As a result, they’ve filed a petition asking the court to dismiss the suit on those exact grounds.
The argument for the banks’ position is: They are completely correct. The law, as reflected in judicial precedents but more importantly in the absolutely universal practice of banks and loan investors, is that syndicated loans are not securities. Everyone who buys a syndicated loan—and again, those people are exclusively large professional investors—knows and agrees that loans are not securities. “Leveraged-Loan Investors Can’t Feign Gullibility” is my colleague Brian Chappatta’s headline about this case.
The argument for the investors’ position is: Well, boy, loans sure are a lot like securities. They are in many respects interchangeable with bonds. If it would be securities fraud to sell bonds without certain disclosures—the claim here is that JPMorgan and other arrangers for a loan to Millennium Health LLC knew about a regulatory investigation and did not disclose it—then it should be equally securities fraud, or at least something-like-securities-fraud, to sell a loan without those disclosures. It’s the same basic product being sold to the same basic group of investors with the same basic sorts of disclosure; why should the outcome be different?
I think that’s a bad argument, but I can see why someone would make it. It’s not wrong. The loans are like bonds. If the rules for bonds are good, then it stands to reason that those same rules might be good for loans. This is not a sufficient argument, though. The fact that two things are similar doesn’t mean that they are regulated similarly, or even that they should be. Having two similar products with two different regulatory regimes gives investors and issuers more flexibility. For instance, the fact that loans are not securities means that they can be sold with less disclosure than bonds—or more. They can be sold both to traditional banks that have traditional bank-type relationships (including doing deep nonpublic due diligence on issuers) and to hedge funds that also trade bonds and want to remain free to trade the issuer’s securities. From the Loan Syndications and Trading Association’s brief in this case (citations omitted):
Unlike bonds, syndicated term loans are originated, syndicated, and traded on the basis of confidential information, which can include material non-public information within the scope of the securities laws. Market participants can choose whether to be on the “private side” and have access to such information, or to be on the “public side” without access to such information, so that they can continue to trade in the borrower’s securities. This means that there will often be disparities in the information available to different lenders. “Public side” participants have “elected to make decisions with respect to a loan without accessing … information available to private side [participants], even though such information may be material to a decision whether to acquire or dispose of such [a] loan.” …
The same is true when loans are transferred on the secondary market. The standard terms and conditions for such transactions provide that each party acknowledges that the other may have information about the borrower that may be material to the decision to enter the transaction; that it has chosen to enter the transaction notwithstanding that it lacks such information; and that no liability attaches to either party for nondisclosure of such information, as long as the representations and warranties in the agreement itself are accurate. In short, the syndicated term loan market operates on the understanding that the parties do not need or expect a mandatory disclosure regime like that applicable to bonds and other securities.
The point here isn’t so much that this regime is good—it has some big benefits, though it also means that you can get burned by trading with someone who has more information than you—but that it’s the regime that everyone agrees on. There is a pretty widespread assumption in financial markets that, if everyone in a particular market is a sophisticated professional, then the rules that they agree on are the rules. Sometimes courts and lawyers come in and disagree, but it’s a decent working assumption.
The art market is the best because it is what you would get if you took an incredibly sophisticated financial market and stripped out (1) fundamentals and (2) liquidity. The value of a financial asset is the discounted present value of its future cash flows, but works of art have no future cash flows—except whatever you eventually sell them for—so their valuation is based purely on sentiment. (Yours, if you like the art, or your estimate of other people’s future sentiment, if you are just a speculator.) Each work of art is unique, or unique-ish anyway (lotta Hirst dots), and none of them trade very often, so, compared to most financial assets, there is just not a lot of data to inform your valuation of any particular work.
So art valuation is, uh, more art than science, sorry. This means that some stuff that happens in financial markets— dealers lying to customers about the price that they paid for a bond, say— happens in comically exaggerated form in the art market. If you buy a bond for 79.50 you can tell a customer that you paid 79.8 and charge him 79.9, but there are some limits; even a relatively unsophisticated customer knows that the bond didn’t cost you, like, 200. Bonds just don’t do that. But you can buy an art for $1,000 and sell it for $450 million, which gives you a lot of leeway to make up a number in between.
But the really delightful thing is that so much of the apparatus of modern scientific finance exists in the art world anyway. So here’s a story about a put-writing strategy in fine art. The strategy is that rich collectors (or financial speculators) agree to write puts to auction houses, guaranteeing that they will buy a work that is going up for auction if it doesn’t sell for more than the guarantee price. Typically the guarantee price will be below the low estimate for the work; the guarantor will be paid a cash fee for the guarantee, and, if the work sells to someone else, the guarantor will get a share of the upside and of the auction house’s commission.
If you have marketed financial derivatives, as I have, all of the beats in this story will feel weirdly familiar. There is put-selling as an income strategy, a way to bet on prices going up without (you hope) actually acquiring the asset:
Increasingly, the goal is to earn a quick payout from the fees surrounding guarantees rather than to bring home any actual paintings.
Though of course you run the risk of the put being exercised against you in a down market:
Only once, about six years ago, did he guarantee a work he didn’t admire but considered a profitable bet. That work is now hanging in his London office.
In addition to the income strategy, there is also put-selling as a tactical way to acquire the asset at an attractive overall price, with the put premium subsidizing your purchase price:
Bidders competing with third-party guarantors need to know the odds may not be stacked in their favor, collectors say. Guarantors essentially get a discount on works because they can glean financing fees or other discounts they can apply to future purchases or shipping. “When the house pays a financing fee on the guarantor’s winning bid, that fee ends up discounting the final price and that creates an uneven playing field for other bidders on the lot,” said Megan Noh, an art lawyer with Pryor Cashman. “The guarantor’s money is essentially cheaper than theirs.”
As with any financial derivative, there’s a paragraph about how someone is trying to standardize and package it and sell it more broadly:
One London firm, Pi-eX, is even trying to turn the guarantee into a derivative, with shares of risk being parceled out to conceivably dozens of investors who don’t know one other, much less share the same taste in art.
There is the perennial story of financial innovation creating lucrative opportunities for first movers that are quickly competed away as the innovation becomes widely known and copied:
Before the recession, the handful of people doing third-party guarantees often reaped 50% of the overage and half the buyer’s premium in exchange for staking a work. Today, guarantors say their expanding ranks have already started whittling typical returns to between 15% and 30% of the overage.
And there is the backlash and return to simplicity, as derivative buyers realize that they’re probably overpaying for the protection they’re getting 10 :
Sotheby’s CEO Tad Smith said on a recent earnings call that some sellers are so confident, they’re bypassing guarantees and simply negotiating some of the house’s own fees on their art’s sale, or a so-called enhanced hammer deal.
It’s basically the entire story of financial derivatives, but about art. I mean, fine, it’s missing some elements. There is no dynamic hedging. No one is selling puts on art and hedging them by shorting a replicating portfolio using the Black-Scholes formula. Give it time, though; eventually someone will guarantee a Damien Hirst shark and hedge the risk by selling a bunch of dots.
Former Top Financial Regulators Warn Against Move to Ease Oversight of Firms. Regulators cut banks some slack on shift away from Libor. Pimco and TCW funds whacked in Morningstar ratings revamp. Chinese boom for hedge funds at risk from US trade dispute. BlueMountain’s Future Is Dimmer After the Hedge Fund Shuttered Its Equity Book. New York Blockchain Week Begins With Fewer Lamborghinis. Inside the More-Is-More World of Selling Luxury Yachts. Canadian rugby team launches range of cannabis-derived products. UEFA Investigators Set to Seek Manchester City’s Ban From Champions League. An Interview With A Man Who Eats Leftover Food From Strangers’ Plates In Restaurants. “I transformed my iPhone into a landline by disabling notifications for every application except calls, and leaving it plugged into a wall outlet in my kitchen.” The Danger Of Drugs: This Man Has Become Annoying After Doing A Drug.
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!
Particularly Uber, the IPO buyers who paid $45, thepre-IPO investors who paid $48.77 over the past few years, and the Uber executives who stand to make a lot of money if the price gets above $120 billion.
For instance: Uber (for not making money), Lyft Inc. (for announcing bad earnings as Uber was going public), public investors (for being short-term-oriented or whatever), private investors (for being overoptimistic), Donald Trump (for starting atrade war), etc.
Total syndicate fees were $106.2 million, or about 1.3% of the deal proceeds. (I had earlier mentioned that Uber’s preliminary prospectus implied a 1% fee; I am not sure how or why that changed.) Morgan Stanley got about 38% of the fees, or about $40.6 million.
That’s a rough guess. The volume-weighted average price between 12 noon and 2 p.m. on Friday, May 10, was $44.0151, according to Bloomberg; about 67 million shares traded. (The IPO opened for trading at about 11:50 a.m.) After that, the stock started declining. It’s unlikely that Morgan Stanley was almost half the volume during that period, but if you assume that (1) they were and (2) they bought at the average price, then that’s about $26.6 million in trading profits. (They sold the 27 million greenshoe shares for $45 each and bought them back for $44.0151 each.) That seems like a floor on their greenshoe profits.
The volume-weighted average price on Monday, May 13, between 9:45 a.m. and 12 noon, was $38.0729. I use 9:45 as a start time because I assume they didn’t do much stabilizing at the open but who knows. If they sold 13.5 million shares (half the greenshoe) at $45 and bought them back at $38.07, then they made about $93.5 million—plus another $13.3 million on the shares they bought back, in this hypothetical, at $44.0151 on Friday.
I mean, a lot of people would say that the mortgage crisis was an exception, in that the banks were securitizing their loans, so they didn’t care about being paid back, so they really did forget to ask questions like “do you have a job?” More generally, a lot of modern bank lending—spammy credit-card offers, for instance—is pretty much a statistical game where they don’t ask many questions at all and expect some level of losses. What I describe in the text is perhaps not a completely accurate description of modern automated consumer lending, but it still gets across the gist of business lending.
This is partially true even inprivatesecurities offerings, where the rules are not quite as strict as they are in public offerings, but where practices still tend to be disclosure-oriented.
On the other hand the arrangers generally lend their own money to the company, which is less commoninsecurities offerings.
In practice this is more likely to involve an online data room with disclosures about the companythan, like, the bank sending loan officers to do a thorough inspection. And the data room will likely respond mostly to questions asked by the lead arrangers, not every question that any lenderhas. But the theory is, you get to do your own due diligence.
That is how Iwould analyze this paragraph: If the auction house is typically spending a lot of its commission to obtain a guarantee, then the commission is part fee-for-service and part put premium, and if sellers don’t want the put theycan negotiate not to pay that part of the commission.