Credit Suisse vacation derivativesSome of the world's best financial engineers work in human resources at Credit Suisse Group AG. In 2008, when the banking system was awash in toxic mortgage assets, Credit Suisse paid its employees with toxic mortgage assets. This was called the "Partner Asset Facility" and it was great for the bank: In a frozen illiquid market where mortgage exposures were viewed with suspicion, it was able to move this stuff off its books. It was great for the employees: "Toxic" assets are fine at the right price, and the employees got them at the lows; they regained a lot of value and the employees made a killing. It was great for aligning incentives: Credit Suisse's employees had to eat what they killed; knowing that they could end up owning Credit Suisse's worst stuff, they would make sure that stuff was good. In 2011, as financial regulatory reform forced banks to focus more on systemic and counterparty risk, Credit Suisse paid senior employees in credit exposure to its derivatives counterparties. Credit Suisse had a bunch of derivatives, and if its counterparties didn't pay up on those derivatives that would be bad for Credit Suisse, so it handed that risk to its employees, as their bonus. Enjoy! I think that it would not have occurred to most people, or even most investment bankers, that you could pay employees in risk. In a sense it is obvious—what are stock options if not paying employees in the form of risk?—but no one really thinks of it that way; only a few simple, standard risks (mostly: your own stock price) are common currency for employee pay. But Credit Suisse thought: Look, we have cash, and we have stock, but we also have lots of other assets. Every asset, for a bank, is a risk exposure, a possibility that the asset won't actually pay off. Some of our assets are very liquid and easy to value and we should hang on to them. Others are illiquid and hard to value and disfavored by regulation and the market; hanging on to them is expensive, and selling them won't fetch a good price. But if we make our employees take them, in lieu of a cash bonus, we can get a lot more value for them, while also aligning incentives and improving our balance sheet. It's a good trade, an arbitrage really. Doing financial engineering in HR just makes sense. Investment banking is a business where the assets walk out the door every night, the cliché goes, and while this is not quite true—the assets of a modern universal bank are stocks and bonds and loans, and they don't walk anywhere—it does get at something real. The rule of thumb is that roughly half of a bank's revenue will go to compensation expense; for Credit Suisse last year the number was about 45%. If you are in the business of financial engineering and looking to expand your market, your employees' compensation is a great place to start. It's a huge market (half your revenue!), they're easy to target and get a meeting with (they work there!), you know they're open to complex financial structuring (they work there!), and they will have a hard time saying no. Most banks' financial engineering of employee pay doesn't go much beyond "we'll pay you more if you make more money and less if you make less money," but really the possibilities are infinite and Credit Suisse, to its credit and my delight, keeps an open mind. Times are tough now, and a little creativity is in order. In 2008, times were tough because the mortgage market had collapsed; banks had too many mortgage assets that they couldn't get rid of, so Credit Suisse cleverly got rid of its mortgage assets by giving them to employees as pay. In 2020, times are tough because the global economy has shut down and people are not supposed to go to work. What do banks have too much of? Well, they have a lot of not going to work. That's pretty abstract, but it's a start. "Not going to work" is not a conventional financial asset, but it clearly has a value; you can do something with it. Maybe you could pay people in the form of not going to work? Maybe you could sell not going to work? Credit Suisse Group AG is offering to sell extra time off to top executives and managers based in Switzerland to give them a personal stake in the bank's cost saving efforts. Switzerland's second-largest bank is recommending higher ranking employees purchase two additional weeks of vacation this year, according to a Credit Suisse spokesman. "This will allow managers to make a personal contribution to a responsible approach to costs," the bank said in a statement. Credit Suisse said other employees could also choose to purchase two more weeks of vacation as part of the employee benefits program.
So … is the price of two weeks' vacation higher than, lower than, or equal to two weeks' salary? Like is this two weeks' partly paid vacation (lower), or two weeks' unpaid vacation (equal), or two weeks' unpaid vacation plus you have to chip in some money to keep the bank going (higher)? That question is complicated by the fact that so much of banking pay comes in the form of the year-end bonus, so "two weeks' salary" is not really your compensation for two weeks of work. Maybe Credit Suisse could run a Dutch auction for the vacation time and sell it for whatever price clears the market. Set up a secondary market! Act as a market maker in the secondary market and collect the bid-ask spread! Sell futures and options on vacation time! Structured notes! Build an entire asset class, really, why not. All it costs you is time, and that's the thing you have too much of. No LuckinA good business model is to make a product that people want and then sell it for more than it costs you to make it. If you do this well—if you sell lots of units, and spend much less on making them than you get for selling them—then you will get rich. But it is hard. You might not be able to figure out what people want, or how to give it to them, or what they want might cost more to manufacture than they are willing to pay. There are other business models. For instance you could make a product that people kind of want, or that they would want if it were affordable. Then you convince people to buy it by selling it for much less than it costs you to make it, or by paying them to buy it. If you do this well, you will have high revenue and rapid revenue growth, because lots of people are buying your product. You will not, however, get rich, because you'll be spending more money making the product than you get from your customers. Your revenue will be high but your net income will be negative; it will cost you money to run this business. But then you will go to investors, and you will say "look, I have a company with rapidly growing revenue, that's worth something, you should pay me for a share of my company." And they will agree—"we love rapid revenue growth," they will say—and you will sell stock in the company for hundreds of millions of dollars. And then it will cost them money to run the business, and you will be rich. There are various possible endgames; in some of them you go to prison but in quite a lot of them you just stay rich and become an elder statesman admired for your business acumen. We talk about this model all the time. One version of it is what is sometimes called the "MoviePass economy," after the company that would give its subscribers unlimited $15 movie tickets for $10 a month. That is a product that people want, sure, in the sense that people who see movies would prefer to pay $10 for unlimited movies instead of $15 for one movie. But it is, comically obviously, not a viable business, because you spend more buying the tickets than you get in revenue. But you can show rapid revenue growth, and you can talk a good game about economies of scale (maybe if everyone signs up for MoviePass it can buy the tickets at a discount?) and network effects and selling user data, and you can convince investors that you have both (1) a lot of customers for your non-viable business and (2) the germ, somewhere, sometime, somehow, of a viable business that you could do with those customers. And that's enough to raise money from investors and have a billion-dollar valuation. Sometimes it all works out: The reason investors will pay for this is that sometimes "flipping the profit switch" works. It is in fact true that having a huge loyal customer base is a good thing for a business, and that spending money to build a huge loyal customer base and then raising prices or cutting expenses or selling can be a good way to get rich. Another version of this model is fraud, though. If your business model is not to sell your product for more than it costs to make it, but just to sell as much of it as possible, the obvious move is to sell a lot of it to yourself. Pay yourself a billion dollars for a billion widgets and you've added a billion dollars of revenue, which is great if investors will pay you a multiple of revenue for your stock. Of course doing this costs you a billion dollars, and you don't have a billion dollars, but there's a simple solution to that. Set up three companies, Company A, Company B, Company C. Company A buys widget components from Company B for $1 billion and sells widgets to Company C for $1 billion. (Or, like, $999,873,164 and $1,000,246,089; just make the numbers look a little real, you know?) You take Company A public at a multibillion-dollar valuation; Company B and Company C stay private, and are effectively just checking accounts that you control. The $1 billion that Company A pays Company B for widget parts goes right back to Company C to pay for the widgets. Ideally you don't even make the widgets, or send any checks; all of this exists purely as a matter of book entries. (We have talked about this version before too; it is easiest to execute if your product is virtual, so you don't even need a widget factory.) Luckin Coffee Inc., the Chinese coffee startup, allegedly executed this model beautifully: China's upstart Luckin Coffee Inc. grew at a blinding pace. It opened stores faster than Starbucks Corp., doubled its valuation to $12 billion eight months after going public and pleased its big-name investors in the U.S. Then, on April 2, Luckin said many of its sales had been faked. … It turns out that Luckin sold vouchers redeemable for tens of millions of cups of coffee to companies that had ties to Luckin's chairman and controlling shareholder, Charles Lu, according to internal documents and public records reviewed by The Wall Street Journal. Their purchases helped the company book sharply higher revenue than its coffee shops produced. Meanwhile, other internal documents showed a procurement employee called Lynn Liang processing more than $140 million of payments for raw materials such as juice, delivery and human-resources services. Ms. Liang was fictitious, according to people familiar with Luckin's business. ... A look at registration records of companies that bought vouchers and others that received repeated supplier payments shows that many had links to Luckin, Mr. Lu or Mr. Lu's two previous ventures. Some listed the same office addresses and contact numbers as branches of CAR Inc. or Ucar [those previous ventures]. Several were registered with email addresses of employees of those companies. One was registered with a Luckin email address. A few of the companies had links to a relative or a friend of Mr. Lu. One regular bulk buyer of coffee vouchers, Date Yingfei (Beijing) Data Technology Development Co. Ltd., has the same phone number as a branch of CAR Inc. and a predecessor of Ucar.
Buy coffee from yourself. (Or juice, HR services, whatever; it doesn't matter, since you're not actually delivering any of it.) Sell coffee to yourself. (In the form of vouchers, since, again, you're not delivering anything.) Report high and growing revenue. Watch your valuation rise. Sell stock. People familiar with these transactions surmised that, over time, the rafts of purchases and payments formed a loop of transactions that allowed the company to inflate sales and expenses with a relatively small amount of capital that circulated in and out of the company's accounts.
It's worth saying a bit about how Luckin got caught. It's especially worth saying a bit about how Luckin didn't get caught. You might expect that Luckin's auditors would catch this trick, which is truly the oldest trick in the book. They did not. Luckin's stock trades in the U.S., but its auditors in China are not subject to U.S. regulatory oversight or examination, and they did not cover themselves in glory here. One result of this is that the U.S. might restrict U.S. listings by Chinese firms, since their accounting is sometimes suspect. How did they get caught? On Jan. 31, Muddy Waters LLC, a U.S. short seller with a record of exposing misbehavior at Chinese companies, circulated an 89-page unattributed report on Luckin. The report said an examination of more than 11,000 hours of video footage of customer comings and goings, of more than 25,000 customer receipts and of observation by 1,500 individuals who visited Luckin outlets showed that much of the company's revenue must be fabricated.
The thing about inflating your revenue by pretending that you sold more coffee than you did is that people can go to your stores and watch you sell coffee. It is a reasonable bet that they won't do that, because it's incredibly boring. "Who is going to send 1,500 people to our stores to watch us sell coffee all day, count how much we sell and compare it to our financial statements," Luckin could reasonably have thought.[1] But the answer was "short sellers"! They actually hired people to sit around watching the coffee get made, so they caught the fraud. It is … almost … a self-limiting system. The more fake coffee you pretend to sell, the higher your valuation will be, and the more money you can make by selling stock. But the higher your valuation is, the more incentive short sellers will have to discover that your coffee is fake. With a few million dollars of revenue, no one is going to send a thousand people to sit in coffee shops and watch the coffee get made. With billions of dollars of rapidly growing revenue, somebody will. Everything is securities fraudIf you run an exchange-traded fund that bets on the price of oil, and the price of oil goes down, is that securities fraud? A popular exchange traded fund that uses complex derivatives to track oil is being investigated by U.S. regulators over whether its risks were properly disclosed to investors, scrutiny triggered by crude's historic slump during the coronavirus crisis, said three people familiar with the matter. The Securities and Exchange Commission and the Commodity Futures Trading Commission have both opened probes into the $4.64 billion United States Oil Fund, which lost 75% of its value in the two months ended April 30, said the people who asked not to be named because the matter is private. Issues the agencies are examining, the people said, include whether shareholders were adequately informed that the ETF's value wouldn't necessarily move in tandem with the spot price of oil and the fund's recent decision to purchase crude contracts that expire further out in the future. The change in contracts USO was buying deviated from its past investment strategy. The inquiries into the ETF, known by its ticker USO, are in their early stages and may not lead to allegations of wrongdoing. United States Commodity Funds, the company that manages USO, hasn't been accused of any misconduct by the SEC or CFTC.
When we first talked about last month's oil-price crash, one thing that I said was that it would be nice if there was a way to buy a financial product that exactly tracked the spot price of oil. "Permanent abstract oil," I called it: "You buy it for the price of oil today and sell it for the price of oil when you want to sell it." A lot of people want to speculate on the price of oil (as USO's size demonstrates), and mostly what they want to speculate on is that. They want a thing that goes up when the price of oil goes up and down when the price of oil goes down; that is the only thing they want to bet on. But there is not exactly a product like that.[2] There are oil futures, but they expire: They have a fixed life, some number of months, and then they end and, if you still own them, you have to take delivery of oil.[3] You can build a perpetual strategy around them—buy next month's futures, wait, and as they get closer to expiry sell them and buy the following month's futures, etc.—but then the thing you are betting on is not quite the thing you want. You are betting on the relative value of different futures, the shape of the curve, the cost of rolling futures, all this technical stuff. "I'd wager that 90 per cent of investors in USO couldn't explain what contango is," a hedge fund manager told Rana Foroohar, but contango is a key part of their actual USO bet. The thing is that oil futures ETFs like USO are obviously an attempt to solve that problem. You can buy USO and hold it for as long as you want; it never (you hope!) expires. It is like a share of stock, a permanent bet on oil prices. On any particular day it will probably go up if oil prices go up and down if oil prices go down, though over the long term it will tend to go down either way.[4] If you want a simple, permanent, indefinite, abstract bet on oil prices, one with a simple name like "U.S. Oil," USO—not July WTI crude futures or whatever—is the product for you. It's just that that product is actually impossible to manufacture; USO is a good-faith approximation, but it doesn't quite get there. Over the long term USO won't really track the spot price of oil, and when things get weird, as they have for USO, it will get even farther away from that simple bet. All of this is disclosed, of course—USO's prospectus explains how it works—but there is still a disconnect. Investors want a simple bet on oil, and USO was built to give them the simplest possible bet on oil, but it's still not as simple as they expected it to be. Are floor traders good?I keep saying that we are in the middle of a big natural experiment to answer that question, but maybe I am just too optimistic. Really we are always in the middle of a natural experiment like that: There are some all-electronic financial exchanges, and some part-human exchanges, and you can always try to compare their performance. But there are various confounding factors, there are different measures of performance that will cut different ways, and there are people on both sides with a vested interest in arguing one way or the other. Now, due to a pandemic, the human exchanges have gone all-electronic, so their performance is different, and you might ask whether it is better or worse. But the same problems will arise and make the answer debatable. Did shutting down CME Group Inc.'s eurodollar pit make trading better or worse? If you ask an algorithm (well, a human who makes algorithms), the answer is of course better: "Nobody was prepared to go to 100% electronic, but the market did it with no problem whatsoever," said Thomas Fitch, founder and CEO of RVAssets, which has supplied trading algorithms for eurodollar and Treasury options on the CME since 2014. He disputes the assertions from floor traders that they provide tighter spreads. The two months in which eurodollar options trading has been wholly electronic allow for a near-complete analysis of more than 90% of trade data, Fitch said, and it reveals an average bid-ask spread of 0.26 cent. The average for floor trading is impossible to measure, but was probably 0.30-0.40 cent, he estimates. And traders at the exchange form "an opaque layer of brokerage" that's able to gather information about flow that a screen doesn't convey, he said. ... The closing of the pits has driven adoption of electronic innovations in development for years and investors "can trade any strategy today as easily as they could prior to closure of floor," Sean Tully, CME's global head of financial products, said on the company's earnings call.
If you ask a pit trader, the answer is worse: Stakeholders in open-outcry dispute that statement. Since the floor closed, illiquidity is particularly acute in weekly options on Treasuries and eurodollars and in long-dated eurodollar structures, said Matthew Carinato, chief operating officer of Trean Group LLC. Carinato said it was likely that some of the options business that used to go to the floor had migrated away from CME's listed products to the decentralized swaps market.
Also: Because of the sheer number of possible combinations when every underlying futures contract, expiration month and strike price is taken into account, human market-makers shouting and flashing hand signals can work faster and at lower cost than robots, according to the humans.
I feel like "quickly doing math on lots of different potential combinations" is actually a specialty of computers? Usually when I read defenses of floor traders they are focused on the humans' calm common sense, not their superior ability to rapidly do complex computations. Things happenU.S. Corporate Bond Sales Smash Record, Soaring Over $1 Trillion. Goldman's Eccentric Couch-Surfing Partner Plans Own Credit Fund. Virus Sparks Round-the-Clock Rush to Fill Gold Vaults. Peet's Coffee Raises $2.5 Billion From IPO, Defying Pandemic. The World's Oldest Exchange Is Europe's Hottest Listing Venue. Mnuchin's $29 Billion Loan Fund Untapped as Airlines Eye Rebound. Turkey Borrowed Foreign Currency as Lira Tumbled. The Contagion of Concern. It's a Tough Time to Be a Fan of Bats. Protestors Criticized For Looting Businesses Without Forming Private Equity Firm First. 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] It's actually not just "compare it to our financial statements." The best part of this story might be that they compared the count of coffee sales to the *receipts*. From the anonymous report that Muddy Waters circulated: "As all orders are placed and paid online and picked up offline, when an order is placed, a three-digit pick-up number and a QR code will be generated to facilitate the in store pick up. Some may have observed that the three-digit pick-up number appears sequential within each store in a day and shared by both pick-up and delivery orders; and use it as a barometer to track the daily order volume of a store by placing orders at both opening and close time of the stores. This method cannot be used if Luckin intentionally jumps and skips numbers during the day to purposely distort the tracking results." But, says the report, they do: Order number 101 will be followed by, say, order number 121, to make it seem like Luckin is getting more orders than it actually is. "Luckin doesn't necessarily have to jump orders to commit fraud – they can simply fabricate more orders in their financial records. However, here is the clever part: company management likely thought about the possibility that more and more investors and data firms were starting to track their order numbers themselves as part of the due diligence process, so 'jumping orders' is a simple way to mislead investors." In other words, the report argues, Luckin *did* expect investors to check up on its sales by ordering a coffee at the beginning of the day, ordering a coffee at the end of the day, and using the difference between the receipt numbers to estimate how many coffees Luckin sold—so it faked those numbers. But it *didn't* expect the investors to sit in the shop *all day*, count coffees, and compare the actual count to the receipt numbers. [2] I mean a physical oil fund would be a bit like that: You give the fund money, it takes the money and buys oil, it puts it in a tank and waits. There'd be no futures expiry and its value would just track the spot price of oil. It would have to pay storage costs, though; unlike a pure financial ETF that holds only abstract easy-to-store assets, it would have large and potentially variable costs of just putting its stuff somewhere. The economics might end up being a lot like those of actual oil-futures funds like USO; arguably part of the roll costs that USO pays in the futures market are, effectively, a way to pay for storage. [3] There are also cash-settled oil futures, which don't require delivery, so they are easier for financial speculators, but they still have finite lives and raise essentially the same practical problems. [4] I mean, see the performance chart here. For a more nuanced view, check out the discussion starting on page 16 of USO's prospectus. Basically if the oil market is in contango for a long time—if next month's contract is more expensive than this month's—then USO will always pay more for next month's contract than it gets for this month's, so over time it will spend down investor money on roll costs even if the spot price of oil drifts up. The easiest way to understand this is to pretend that USO is a physical oil storage fund (see note 2): If investors gave it money and it bought a bunch of oil and put it in tanks, it would have to pay rent on the tanks every month, and that would slowly fritter away investors' money, while the oil in the tanks would bring in no cash flow. The roll costs, during prolonged periods of contango, are effectively those storage costs. | 
Post a Comment