The Blood Unicorn Is Extinct
(Bloomberg Opinion) -- Theranos.
What would happen, do you think, if Theranos Inc. had announced today that it had built a working blood-testing machine that could perform a broad range of common blood tests on a finger-prick sample of blood? You know, like it said it had done years ago, only for real this time? What if the story of Theranos, the Blood Unicorn (Elasmotherium haimatos), was that it was working hard and fast to build a real revolutionary blood-testing device, and it missed some deadlines in that process, and out of defensiveness it cut corners and lied to investors and patients about its progress, and the Securities and Exchange Commission accused Theranos of securities fraud, and federal prosecutors accused founder Elizabeth Holmes and former president Sunny Balwani of wire fraud, and the company was battered and reduced to a shell of its former self, but its resolute engineers just plugged away at their blood-testing device, and today—after the fraud charges and indictment and terrible publicity and all the rest—they announced that they had cracked it, and that the revolutionary promise of Theranos had been fulfilled?
Would that … work? Would everyone be like “never mind, high five, welcome back to polite society”? Not really, right? Lying to investors about material scientific and financial results is fraud, even if the science eventually works out in the end; lying to patients about the blood tests they’re taking is even worse. Still it would make the story quite a bit more sympathetic. The revolutionary inventor whose single-minded pursuit of her ambitious vision causes her to ruthlessly step over anyone in her way, and who eventually achieves that vision, is a Silicon Valley archetype. “Her only sin was that she cared too much, about the blood testing,” her lawyer would tell a jury. But just ruthlessly stepping over anyone in your way, and going nowhere: That’s a bad look. Doing instrumental fraud in pursuit of a real vision is practically the American way. Doing pure fraud—fraud in the service of fraud, fraud with no reality behind it—is just bad.
Anyway no Theranos did not announce today that it had built a working blood-testing device. Instead, the Wall Street Journal reported that it will shut down, a result that is deeply unsurprising but also a bit disappointing, in that it cuts off any real possibility—if there was any real possibility—that somehow this story could end in redemption, and blood tests. Here is the email that its current chief executive officer and general counsel sent out to investors, explaining that it tried and failed to find a buyer, is in default on the loan that Fortress Investment Group gave it last year, and will now be dissolved. There will be a little money left for creditors, perhaps $5 million for $60 million of unsecured debt. Shareholders will, of course, be zeroed, though at least they got a wild story out of their hundreds of millions of wasted dollars. Fortress will get Theranos’s intellectual property. Maybe they’ll build a blood-testing device with it!
Yesterday I wrote about the idea of letting public companies report earnings every six months instead of every three months. This was in the context of a story about how miserable earnings season is for analysts and fundamental investors, and how much more time they’d have to eat and sleep and see their families if they only had to do it twice a year. “Pretty much everyone involved with public companies would benefit, from a lifestyle perspective,” I wrote.
A couple of readers objected: Well, if there was half as much work, wouldn’t banks employ half as many analysts? Sure the analysts would get to spend more time with their families, but only because they’d be laid off for lack of work.
This is, I think, a bit of a strange objection. The job of a stock analyst is not primarily to process earnings releases. The job of a stock analyst is primarily to help investors understand companies. They do this by researching the company, writing reports, talking to company executives, talking to investors, setting up meetings between investors and executives, etc. One aspect of this is that when a company puts out a press release saying “we made $1.50 per share last quarter,” the analyst will put out a note saying “it is good that this company made $1.50 per share last quarter” or whatever, but that is not the highest-value-add thing that the analyst does. Lots of other investors and analysts and algorithms are reading the earnings release; the news in the earnings release gets incorporated into the market fairly quickly. The analyst is most valuable when she is telling investors something they don’t already know.
One way to think about it is that analysts’ jobs are currently precarious because they face unfair competition from companies. The analyst’s job is to help investors understand companies, to write reports predicting how much money the companies will make, etc. When the companies put out their own detailed press releases just saying how much money they’ve made, when they give guidance about how much money they’ll make in the future, when they hold public calls and take questions to help people understand their business, they are doing for free what the analyst is paid to do. (Well, the analysts are paid pretty indirectly, but never mind that.) If they stopped doing that, then there would be more demand for analysts. If companies just tell the public everything that it needs to know, then what value does an analyst add? But if companies are opaque and mysterious, then someone who has researched them deeply, who talks regularly to executives, who can set up one-on-one investor meetings with those executives, who has built an informed model of their future earnings, etc., serves a vital purpose.
It seems to me that the point of having lots of mandatory public disclosure is to create what regulators like to call a “level playing field.” This appears to mean that you don’t have to be a well-connected experienced financial-industry professional to figure out what stocks to buy. You can, for instance, be a day-trading hobbyist: You can read a company’s financial reports and draw your own reasonably informed conclusions about its business. But you could also be an algorithm created by a couple of smart computer scientists that uses voluminous public data to predict price movements, without ever bothering to set up one-on-one meetings with corporate executives. In a world of perfect instantaneous total public disclosure—where every relevant fact about a company’s business was disclosed immediately—amateur traders would probably have a tough time of it, because presumably they have day jobs and wouldn’t be able to read the disclosure constantly. But algorithmic traders would make a killing, because their skill set is processing lots of data quickly, and professional fundamental investors would be left behind.
But in a world of infrequent minimalist disclosure—say, where companies report earnings every six months and don’t give guidance or do public earnings calls—it seems to me that the fundamental professionals would gain an advantage. The playing field would be less level for the amateurs, and there’d be less data for the quants. The people who bother to do the detective work of digging into companies, and the human work of sitting in meetings with their executives, would be relatively better off. There is a reason that quantitative investing—and retail investing, for that matter—is prominent in public markets but relatively absent in private ones.
I don’t mean to suggest that this is a good thing! It’s just a thing, a set of tradeoffs, some people winning and some people losing. You might quite reasonably prefer the level playing field. You might think that the (already extremely common and apparently legal) one-on-one meetings between investors and companies are not the best (fairest, most efficient, etc.) way to disseminate corporate information to the market. But if your business is knowing companies better than anyone else, a change that makes it harder for other people to know companies—that raises the barriers to entry into the knowing-companies business—is probably good for you.
I have a certain perverse fondness for “last look.” The idea is that a market maker quotes a market on a thing—offering to buy it for $100.10 or sell it at $100.15—and if you come to her and say “yes okay I will buy it from you at $100.15,” she gets a brief chance to say “never mind.” If the price has moved against her in that brief delay—if now the thing is trading at $100.25 or whatever—then she doesn’t have to do the trade at $100.15. On the other hand if the price has moved in her favor—if now it’s trading at $99.95—then she cheerfully executes your trade at $100.15. This is obviously good for the market maker; I once wrote:
It is as perfect an embodiment of "heads I win, tails you lose" as you could ask for: If the price moves against the customer, the bank wins; if the price moves against the bank, the bank decides not to play.
There are, I think, two things you can do with that model. One is, you can stop there, and say that last look is good for market makers and bad for their customers, that it’s a way for banks and other dealers to extract value from investors, and that it’s basically unfair and inefficient and should be banned.
The other is, you can assume that financial markets are fairly competitive, and that in a market with last look—one where market makers have the opportunity to avoid adverse selection by getting away from trades that immediately move against them—the value that market makers extract from last look will be returned to investors in other ways. Specifically it will be returned to investors in the form of tighter spreads. Perhaps in a market without last look the market maker would quote at $100.05/$100.20, knowing that if she sells at $100.20 there’s a good chance the market is going even higher. But in a market with last look she’ll quote at $100.10/$100.15, knowing that she’ll only sell at $100.15 if the market doesn’t seem to be going higher. So investors get to buy at $100.15 rather than $100.20, saving five cents. This benefit can be a little illusory—they only get to buy at $100.15 if the price is stable or going down—but it’s not nothing. If you want to buy a relatively small quantity of the thing in a relatively quiet market, tighter spreads with last look really should save you money.
Last look, as a feature of market structure, is most (in)famous in the foreign exchange markets, and it has lost popularity there as … basically, as investors and regulators and the press have noticed it. It is not a standard feature of U.S. equity market structure. But since IEX got approval from the Securities and Exchange Commission to run a national stock exchange with a “speed bump” delaying execution of orders, other exchanges have pondered building something last-look-ish into the stock market. Here’s one from Cboe Global Markets Inc.’s EDGA exchange:
The speed bump under discussion at Cboe would differ in several ways from IEX’s, according to people briefed on the plan. IEX delays orders to trade stocks by 350 millionths of a second. Cboe’s speed bump would be around 10 times longer in duration, lasting three to four milliseconds, these people said. A millisecond is a thousandth of a second.
The delay wouldn’t apply to all orders equally: Instead, it would affect only orders seeking to hit unexecuted buy or sell orders already posted on EDGA, the people said. Traders posting new orders to be displayed on EDGA wouldn’t be affected.
Such a design would benefit market makers, the firms that facilitate trading by continuously quoting prices for stocks. Market makers would be able to cancel or adjust their quotes without having to wait several milliseconds.
It’s not really a last look: Market makers don’t get to decide whether to accept or reject an order. But it gives them a somewhat similar advantage: They get to see four milliseconds (an eternity!) into the future and see if the stock went up or down. People can only buy from them, at the markets that they posted, if the stock didn’t move against the market maker in that interval. And so in theory the market makers can quote tighter spreads: You’ll be more confident quoting $100.10/$100.15 if you know that you will never sell at $100.15 and then immediately see the stock tick up to $100.25. These tighter spreads are good for investors, though they’re not that good, as frequently the investors won’t get filled at the tight spreads because the market has moved in their favor. (And are more likely to get filled when the market has moved against them.)
Anyway, like I said, I have a perverse fondness for this stuff, not because I think it is good exactly—it does seem a bit sneaky, really—but because I am intrigued by experimentation and diversity in how markets work. (We talked in 2016 about a similar last-look-ish speed bump proposed by the Chicago Stock Exchange, though that was ultimately withdrawn.) If some markets want to have real quotes that market makers have to stick to—so that investors know that if they see a price on the screen they can buy at that price—then that’s great. If other markets want to give market makers some time to back away from their quotes—so the market makers don’t get run over and can quote tighter spreads—then, sure, that’s great too. Give people choices, see what works best for them, why not?
But in the U.S. equity markets those choices are often forced, because the SEC’s Regulation NMS requires brokers to send orders to whatever exchange is quoting the best price. If you don’t like IEX’s speed bump—if you think that it makes execution on IEX too uncertain—but IEX is quoting the best price on the stock you are trying to buy, then you have to route to IEX and deal with the speed bump. It is controversial! EDGA’s much longer speed bump should be that much more controversial, except that EDGA doesn’t want people to be required to route to it:
Under its speed-bump plan, Cboe wouldn’t seek to benefit from an SEC rule that generally helps small exchanges, people briefed on the plan said. The rule lets exchanges offer “protected quotes”—meaning brokers must route customer orders to an exchange if its quote represents the best price for a particular stock. But for quotes to be protected, they must be “immediately accessible,” the SEC has said. That requirement has tripped up other exchanges trying to implement speed bumps.
Cboe wouldn’t seek protected-quote status for orders displayed on EDGA, the people said.
That is something you don’t see every day, a registered national stock exchange asking not to have protected quotes. I am not sure this particular speed bump is a great idea, and I’m not sure how serious it is (“Cboe has yet to file its proposal with the agency”), but part of me is rooting for it for that reason alone. The basic problem with building weird features into a U.S. stock exchange is that the structure of U.S. stock markets sort of forces everyone to use every exchange, so every SEC-approved feature is forced on everyone. (And so the SEC has to be very conservative about approving weird features.) But if you get rid of that problem—if you make your exchange optional—then you could, in theory, make it as weird as you want.
In 2016 we talked a lot in this space about bowls. "Hot, skinny people” were “ditching salads for ‘power bowls’”; they would “always be seeing new bowls and tagging each other in photos,” and swore that “Even if I had the option to eat off a plate, I would eat out of a bowl.” “Plates inhibit you because food slides off,” said a bowl marketer, while a tableware designer explained that “a bowl is much more flexible and open to interpretation compared to a plate.” Even as late as this April I was reading that the bowl “evokes a way of eating that for me is more about something of comfort and sustenance.”
But now bowls are over and there’s this:
The chef, Moshe Segev, says he also serves sweet-cream mushroom linguine on a wood stump, cheesecakes on a painter’s palette and a potato-and-goat-cheese gnocchi on top of a tea pot. “This is not only food, this is art,” he says. “Real art is never 100% for everyone.”
El Ideas, a Michelin-star-rated Chicago restaurant, serves a coconut-and-lime powder on a mirror with razorblades for a 1980s “Miami Vice” feel. Diners suck up the powder with straws.
And the San Francisco version:
Plating took a high-tech turn at Quince, a Michelin-star-rated San Francisco restaurant, which served truffles on iPads in 2016 for the start of truffle season. Screens played videos of a dog or pig searching for truffles. “It looked very foresty,” says Quince beverage director and former general manager Matt Cirne.
And a lot of food in shoes for some reason. Of course the bowl trend was about mass food-marketing sociology, about fast-casual chains and tableware sold at Walmart, while this is about high-concept plating at fancy restaurants. But plenty of trends start high-end and end up everywhere. Look out for suburban fast-casual chains selling poke out of shoes by 2020.
Amazon Hits $1 Trillion Valuation. In A New Email, Elon Musk Accused A Cave Rescuer Of Being A “Child Rapist” And Said He “Hopes” There's A Lawsuit. London’s Tribes of Finance Face Off. Top Goldman Banker Awaits His Fate as New CEO Fills No. 2 Post. Goldman's Gender Bias Fight Moves Closer to Trial After 13 Years. Bitcoin Drops 3% in 10 Minutes, Ethereum Plunges 12%. Future Considerations: Why Ex-MLB Pitcher Michael Schwimer Is Investing in Minor League Longshots. Cohn Lifted Papers Off Trump’s Desk to Stop Nafta Exit, Book Says.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Matt Levine is a Bloomberg Opinion columnist covering finance. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz, and a clerk for the U.S. Court of Appeals for the 3rd Circuit.
©2018 Bloomberg L.P.