Header Ads

Money Stuff: Dorm-Room Hedge Fund Had Its Own Values

Money Stuff

BloombergOpinion

Money Stuff

Matt Levine

Structured notes

A structured note is a bond issued by an investment bank that contains one or more derivatives on some stock or index or other financial quantity. A classic is: You give the bank $100, and in three years if the S&P 500 Index is up you get your $100 back with a return that is some multiple of the return on the S&P 500; if the index is down you get back your $100 with no interest. In practice a lot of structured notes are a lot more complicated than that, but that one is fine for our purposes.

Analytically the structured note is (1) a credit instrument of the bank plus (2) one or more derivatives. The S&P one, for instance, is just a three-year zero-coupon $100 bond of the bank plus a call option on the S&P 500. Those things are pretty well understood and you can—and the bank surely will—plug the parameters of the structured note into a fairly standard set of models and get back a value for the structured note.[1] The value will be, like, $97. The bank will sell you the note for $100. It makes $3. It's a pretty good business, for the bank.

What are you doing, though? The answer can't be "buying $97 for $100." The structured note tells a story, and you are buying that story. With the S&P one, the story is like "you get the upside on the stock market over the next three years, but if the S&P does poorly then at least you get your money back." You get upside with no downside, it is great, where do I sign up.

Cynics can poke holes in this pitch. For one thing, you do have downside risk: The note is generally an unsecured obligation of the investment bank, so it has some credit risk; if the market crashes and the investment bank goes bust, you won't get your $100 back. For another thing, you could probably recreate this profile by (1) putting most of your money in Treasuries and (2) spending some of it to buy call options on the S&P 500 yourself; this might be more efficient because you wouldn't be paying the bank $3 for the packaging. (On the other hand it might not be more efficient, because the bank can probably buy those call options more cheaply than you can, and anyway the packaging and convenience probably are worth something to you.)

But here I do not want to be cynical.[2] Here I want to suggest that the fundamental yet paradoxical business of an investment bank is to spend $97 manufacturing a thing that it can sell to you for $100 and that you value at $115. The bank can source the components of the structured note for $97, because it is in that business and has the knowledge and connections and efficiency to find those components for that price. It can sell the structured note to you for $100, adding a markup for its service. And you buy it because you think it's worth more than $100, because exposure to stock-market upside with limited downside (or whatever the story of the note is) is something that you want and value.

On the one hand this is exactly what every business does: Apple Inc. buys components for $400, puts them together and sells you an iPhone for $1,000, and you wouldn't give up your phone for $10,000. Everyone wins. On the other hand this totally normal business process feels weird and wrong in financial markets, just because the only thing you are ever trading is sets of future cash flows, and everything has a more or less transparent market-based price, and it is hard to see how everyone could win. You don't exactly get $200 worth of psychic pleasure from buying a $100 structured note; you just get back $100 or more or less in three years. If you get back more than you would have gotten by investing elsewhere, it is good; if you get back less it is bad; when it's all over it is hard to see how the thing was worth more to you than its components.

But not that hard. If you buy the note and the S&P goes up and you get back more money than you put in (though less than you would have by just buying the S&P), you will be happy. Sure you could have made more money doing something else, but this way you made money and protected your downside. Similarly, if the S&P goes down you might be happy that you bought the structured note and avoided that downside. The thing worked as intended; it gave you the experience that you wanted; comparing it to some other possible investment sort of misses the point. You are happy, and who would begrudge the bank a profit for making you happy?

Anyway here is a joyfully weird Securities and Exchange Commission enforcement action against a hedge fund that bought a lot of structured notes. "Hedge Fund Started in College Dorm Room Accused of Fraud by SEC," is the Bloomberg headline: The hedge fund is called SBB Research, it was started by a guy named Samuel Barnett in his Caltech dorm room in 2010, and it initially ran "vehicles for investing the Barnett family fortune" before raising money from outside investors.

"SBB invested almost all of the Funds' money in a single asset class: structured notes," says the SEC, and the basic problem is that it allegedly valued the notes wrong:

Although Barnett and [Chief Operating Officer Matthew] Aven had never worked for a hedge fund or created a valuation model, they thought that they knew better than the rest of the market. Starting in 2011, Barnett, Aven, and SBB rejected over 50 years of standard valuation principles, ignored expert advice, and created a home-brewed valuation model that radically departed from the norm (the "Model").

Some inputs in SBB's Model acted like a financial steroid – artificially pumping up note values. Other inputs acted like a masking agent – smoothing artificial gains by spreading them over the notes' multi-year term. SBB created those critical inputs out of whole cloth. None of the Model's "innovations" reflected the assumptions of market participants or were validated by published academic research. Instead, the Model was designed to get results that dovetailed with the Defendants' subjective "intuition" regarding how the notes should be valued. Conveniently, their "intuition" led to consistently higher note values than those yielded by more traditional models.

These inflated valuations, says the SEC, allowed SBB to market itself to investors with inflated returns and charge higher incentive fees. SBB denies the fraud charges and is "eager" to fight them in court—"SBB operates with the utmost integrity and vehemently disagrees with the SEC's aggressive and extreme interpretation of SBB's good faith estimates of their hard-to-value investments," says its lawyer—but I am not going to talk about the fraud aspect because it is not particularly interesting. Instead I want to talk about that model. The SEC says:

First, in valuing the option component of the structured notes, Barnett, Aven, and SBB created a drift term – identified by the Greek letter µ ("mu") – that abandoned the first half of the Black-Scholes-Merton risk-neutral framework. Barnett and Aven concluded that the risk free rate was too low and that using it would yield much lower note values than they wanted. So, they replaced the risk-free rate with the average growth rate for the underlying stock index (a rate often several times higher than the risk-free rate).

Intuitively what this says is that, if they bought a structured note linked to the S&P, and they expected the S&P to go up by 8% a year, then they plugged that growth rate in and figured the note would go up by 8% a year. This is not how option valuation works, for deep and Nobel-Prize-winning reasons, but it is not exactly wrong.[3] In fact it is something like the point of structured notes: The banks sell them because they can manufacture them out of risk-neutral Black-Scholes-y components, but you buy them because they give you upside on stocks and you expect stocks to go up. 

Or:

SBB failed to discount the note to account for the credit risk of the issuer. SBB's Model assumed that – over the multi-year term of the structured note – there was no chance that the issuing financial institution would encounter financial distress affecting its ability to pay. 

Again, wrong! Again, understandable, and again the point of the structured note: The banks sell them because they provide cheap funding, but you buy them because you want the payoff profile described in the note and are pretty sure the bank will be good for it. If you didn't think the bank would be good for it, you wouldn't buy it. This isn't a sophisticated mathematical analysis; you are not discounting the value of the note by a credit spread reflecting the probability that the bank will default. You just hear the story of the note, and believe it, and buy it.

What is strange and lovely about the SEC's telling of SBB's structured-note model is that, underneath the math and the Greek letters, SBB's model recreated the customer's intuition about why you might buy a structured note.[4] "If the stock market goes up I will make money," the customer thinks, and SBB modeled the option using its real expected stock-market growth, "and if the market goes down I won't lose money," the customer thinks, and SBB modeled the downside protection as a true guarantee. Investment banks value structured notes based on standard models that reflect the risk-neutral cost of manufacturing the notes, but customers intuitively value them based on what they expect stock prices to do and how much they trust the banks. SBB just formalized that intuitive valuation.

Obviously those different methods lead to different prices:

The SBB Model's day-one note valuations were almost always higher than the note's nominal (or "par") value. Usually, a structured note's initial value at purchase is less than the note's par value because of transaction costs the issuer incurs creating and hedging the structured notes – costs ultimately borne by the buyer (the "Day-One Discount"). Issuers typically disclosed this reality, and Barnett and Aven knew that such costs were baked into the purchase price. But, SBB's Model consistently produced day-one valuations that were at or above par.

In one extreme instance, SBB bought a note on December 28, 2012 that had a par value of $2,000,000. On the next trading day, SBB valued the note at $2,301,651 – more than 15% over par.

The SEC thinks this is evidence of fraud,[5] and to be clear I am not saying that's wrong. (If you are managing money for investors you have to account for your assets based on market value, not your own optimistic views, etc.) I'm just saying that this accounting reflects something real about how the world ought to work: A bank sells a customer a thing for $100, and the bank concludes that the thing was worth $97 and it made a $3 profit (the "Day-One Discount"), while the customer is sure that the thing is worth $115 and it got more than its money's worth. If the customers buying structured notes didn't regularly think that, they wouldn't buy structured notes. The banks sold structured notes with a story, and SBB's model accurately reflected that story, even if it didn't quite reflect their actual value.

Disrupting the IPO

An initial public offering is supposed to be exciting. It is a big milestone in the life of a young company; after years of toiling in anonymity, you introduce yourself to the public markets and mature as a company and raise a lot of capital to fund the next stage of your growth. You get to ring a bell. 

That is all less true than it used to be. IPO companies are not as young as they used to be, or as small, and the IPO is less important to them. If a company has raised billions of dollars at an 11-digit valuation in the private markets, if it is a ubiquitous global consumer brand run by a famous billionaire, the IPO will just be another day at the office. It's fine, it's a nice thing, you get to ring the bell, but it's not the rite of passage it once was.

You know all this. There is another side of it. An IPO is supposed to be exciting for investors too, just as a marketing matter. Here you are, a small unknown company whose stock has never traded before, and you are trying to get big institutional money managers to do the work to understand your company, dig into your financials, value your stock and take a risk by buying it. You have to make it fun for them, make it worth their while. They don't enjoy the bell-ringing as much as you do.

Really they are boring people, and they mostly like money, and the way you make it fun for them is by selling them stock that will probably go up. But it is hard to credibly signal to them that you will do that, because the whole point here is that they haven't yet met with your executives and dug into your financials, so they have no idea what your stock is worth and whether it will go up. You can't be like "hey, great deal, if you are willing to put in a big order in our IPO, we will sell you the stock for the low low price of $26," because they have absolutely no idea if the stock is worth $26 or $36 or $16.

There is a simple and well-known solution to this problem. What you do is you go to a big investment bank and hire it to lead your IPO. And then the investment bank goes out to the investors and says "hey, great deal, if you are willing to put in a big order in this IPO, we will sell you the stock for the low low price of $26." And the investors look at the last 20 deals that the bank did, and notice they all went up by 20% on the first day, and put in a big order in this IPO. I mean it's not quite that simple—what actually happens is that the investors do the work and meet the company and haggle over the price with the investment bank—but that really is the basic idea. The basic idea is that investors get excited for your random no-name IPO, and put in the work and take the risk, because they have an expectation that most of the time, most IPOs led by most brand-name investment banks go up a lot on the first day.

This is called the "IPO pop." Not every IPO pops, but a lot do; it is generally considered a good and successful IPO if the stock trades up 10 or 20% on its first day. If that is the rule of thumb then IPOs as a general product category will be popular, and investors will clamor to invest in them, and the next no-name company that wants to do one can expect a warm reception.

That is also less important than it used to be. A lot of people dug into WeWork's IPO prospectus for free, not because they expected to make a quick killing on it but because they wanted to make fun of it on Twitter. The excitement was palpable and not always economic. Slack, Spotify, Lyft, Uber, these are companies that were already household names when they went public, and that could fairly expect big investors to do the work to understand and value their stocks. I don't mean "blindly pour money into their stocks"; in fact, public investors did the work on WeWork's IPO and laughed it out of the room. I just mean that investors will give attention and care to big famous names like this. Investors didn't go to roadshow meetings for the Uber IPO because Uber's bankers had a reputation for pricing IPOs attractively; they went to the meetings because it was Uber.

In fact Spotify and Slack didn't go public through IPOs at all; they did direct listings in which their shares just started trading on the stock exchange without any bank negotiated IPO price. Uber closed below its IPO price on its first day of trading; Lyft initially traded up but cracked the next week; both are well below their IPO prices now. A lot of the excitement, for investors, has been sucked out of recent tech-unicorn mega-IPOs. Part of that is about investors' specific reactions to those companies' specific businesses, but part of it is also that you just don't need that much excitement to sell an Uber IPO.

Anyway:

Venture capitalists and executives from hundreds of private companies will meet in Silicon Valley on Tuesday to discuss whether the financial industry's system for initial public offerings is still working after a year in which many of the biggest deals flopped. Attendees plan to discuss alternative strategies including direct listings, which replace financial underwriters with cutting-edge computer code.

"I'm not anti-banker, I'm pro-algorithm," Bill Gurley, a general partner at venture capital titan Benchmark who is one of the meeting's organizers, said in a phone interview. He said investment bankers are largely uninvited.

For months, he and others including Sequoia Capital's Mike Moritz have been pitching the benefits of direct listings, in which computers shift privately held shares to public markets without banks buying giant blocks of stock and parceling them out to clients at a single price the night beforehand. Humans, Gurley said, have been systematically "mispricing" IPOs for decades. 

I mean, if your concern about IPOs is that bankers regularly underprice them to hand huge risk-free gains to their investor buddies, then today is kind of awkward timing for this shindig, no? Gurley's firm took out about $255 million from Uber in its IPO; the shares it sold are worth about $175 million today. SmileDirectClub Inc. and Peloton Interactive Inc. priced IPOs this month and closed down 27.5% and 11.2%, respectively, on their first days of trading. Judging only by recent high-profile venture-backed IPOs, you might conclude that the IPO market systematically overprices IPOs to extract money from public investors and enrich venture capitalists. That is not fair—statistically Gurley is right and most IPOs do pop—but it at least seems like less of a problem than it used to be.

But, sure, right, significant overpricing of IPOs isn't great either. The "pro-algorithm" approach is just, like, you do an auction for the stock and you sell it at the market-clearing price. That's how direct listings basically work (only it's a two-sided auction in which the pre-IPO shareholders decide whether or not to sell), and there's some precedent (Google) for auction-based IPOs too. (There is also some chatter about the possibility of one day doing a direct listing in which the company sells stock, combining the price discovery of a direct listing with the capital raising of an IPO.) It is not traditional, but it is becoming more traditional, and it seems … fine? Here, look, Airbnb Inc. is going to do it:

San Francisco-based Airbnb is laying the groundwork for a direct listing rather than an initial public offering, according to people familiar with the matter who asked not to be named discussing private information. Airbnb declined to comment.

I can't exactly disagree with anything here. It seems to me that the IPO ecosystem is basically good for venture capitalists; it generally lets them cash out of their investments at much higher prices than they paid for them. But it is not clear that the current process is perfectly suited to the current state of the IPO market, and I can see why venture capitalists want to tweak it. I do wonder, though, if there might be some unintended consequences, if extracting some more money from the biggest-name IPOs will make it harder for the smaller ones to get done.

WeKeepGoingWithThis

Well, some people still want IPOs to be exciting. Co-founders Adam and Rebekah Neumann of the We Co., formerly WeWork, for instance:

An S-1 is meant to be a bland financial document, but WeWork's took a different direction. With Adam's encouragement, Rebekah became unusually involved in the artistic presentation of the document. "The traditional approach to producing an S-1 is bankers and lawyers hashing this out, but the process was continually usurped by Rebekah's involvement," one executive said, echoing a sentiment expressed by multiple people who worked on the project. "She treated it like it was the September issue of Vogue." WeWork had hired a former director of photography at Vanity Fair, and Rebekah insisted on selecting the photographers chosen to take photos of WeWork offices and members, and approved every photo that appeared in the S-1, of which WeWork included many more than most companies that go public. … "The thing that's so damning about all that is that it's just not the point of the document," a person who worked on the project said. "That's the thing about WeWork: You're spending all this time working on the surface of it instead of the actual truth of the thing."

I dunno, if your pitch is "we rent offices and lose money, but the offices are really nice," it does make sense to spring for good photography? That is from a New York Magazine story by Reeves Wiedeman about the chaos at WeWork around its (now-cancelled) IPO, which is fun throughout. It also explicitly blames WeWork's investors, particularly Masayoshi Son's SoftBank Group Corp., for the company's problems:

Neumann had always been obstinate, and his company was aggressive and a bit undisciplined, but its business had been on a more sustainable path until Son and Softbank showed up in 2017 and told Neumann to make WeWork "ten times bigger than your original plan," that being crazy is better than being smart, and that WeWork wasn't being "crazy enough." "They're trying to make this all about, 'Adam is a lunatic,'" one rival real-estate executive told me. "But these people invested, they knew the terms, they knew about the governance issues, and they told this guy, 'Be you, but be ten times you.' What did they expect?"

We have talked before about how entrepreneurs respond to incentives, and how, if venture capitalists want to put more money into bad ideas than good ones, those bad ideas will get built. On this model, Adam Neumann was just providing a valuable service and meeting a market need, only the need wasn't startups' demand for for flexible workspace, it was Masayoshi Son's demand for fast-growing money-losing ventures. He saw the opportunity, he built the product to meet it, and now he is dynastically wealthy even if the business never makes money.[6] The botched IPO is a cautionary tale for venture capitalists, yes, but for entrepreneurs this still seems like a pretty attractive precedent.

Things happen

Rep. Chris Collins Resigns Before Expected Guilty Plea in Insider Trading Case. (Earlier.) Fed wrestles with role of regulation in repo squeeze. Venezuela Creditor Cleared to Resume Citgo Seizure Efforts. Argentina sifts through 'fantasy land' bond proposals. Aramco Steps Up Investor Push With Plans for $75 Billion Yearly Dividend. Credit Suisse COO Resigns Over Spying Scandal. Consultant Linked to Surveillance of Former Credit Suisse Executive Reported Dead. Cryptocurrency Exchanges Including Coinbase Disclose Ratings of Digital Assets. An Oral History of Lilith Fair. "After all, what were the chances that the bison would charge again?"

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] If the structured note is a lot more complicated, with best-of options and knock-ins and so forth, the free Black-Scholes calculators you find online will not really give you a value, and you will be entirely at the mercy of the bank to price it.

[2] I have been plenty cynical about structured notes in the past, and I will be again, and the Securities and Exchange Commission has often been cynical about them. It's a complicated product often pitched to retail and retail-ish investors, it attracts cynicism.

[3] It's wrong! It's wrong! Don't email me that it's wrong, I know it's wrong, by no-arbitrage you can manufacture the option using the risk-neutral inputs and so using this mechanism—especially what SBB did, using a risky equity growth rate but risk-free discounting—causes you to overpay. I just mean that it is an intuitive model of the actual expected payoff of the option for an unhedged buyer. 

[4] Well, here's one awkward bit: "Defendants used a smoothing mechanism that spread gains and losses over the term of the structured note (which SBB referred to as 'linearization')." The problem is that by immediately valuing the notes in a customer-y way they immediately got much higher values than they paid for them. So instead of immediately marking them to those values, they just spread their assumed gains out over the full life of the notes. 

[5] "That SBB disregarded the Day-One Discount and at times recorded wildly unrealistic gains in the first days after purchase reflects that SBB was not generating good faith 'fair values' for the Funds' notes. "

[6] Wiedman writes: "Neumann is no longer CEO, but he still has his role on WeWork's board, the homes in Gramercy and the Hamptons, plus a townhouse in the Village, a 60-acre estate in Westchester County, and a $21-million residence in the Bay Area complete with room shaped like a guitar. Neumann had a spectacularly embarrassing summer, but setting aside whatever lawsuits or investigations that could emerge, he's set for life." And who would sue? No public investors bought stock, and like the real-estate guy said, Son knew what he was getting into.


No comments