Maybe Lying About Bond Prices Is Fine
Also back doors, cartels, call spreads and bathroom doors.
We have talked a number of times about federal prosecutors’ crackdown on bond traders who lie to their customers, and I have often described what I think is the best theory of that crackdown, which is:
- It was widely accepted in the more opaque bond markets (residential mortgage-backed securities, etc.) that bank traders and salespeople might lie to customers about the prices the bank had paid for bonds, and that customers should not rely on everything the traders said, and that if the customers caught the traders in a lie they should complain to their supervisors or put the bank in the “penalty box” for a while, but that those lies were not necessarily illegal.
- Prosecutors noticed this and decided that those lies should be illegal.
- Prosecutors had no ability to prospectively make those lies illegal: Federal prosecutors can’t write securities-trading rules, and the people who do write securities-trading rules (the Securities and Exchange Commission) don’t seem to be all that interested in writing new no-bond-lying rules.
- But prosecutors do have the ability to restrospectively make those lies illegal: They can charge some traders who did the lying with “securities fraud,” a broad category, and if they can convince a jury of the quite intuitive notion that “lying to customers to get them to pay you more for bonds is fraud,” then those lies will have been fraud.
- This has the effect of making bond lying prospectively illegal: Once you start sending traders to prison for lying about bond prices, everyone will stop doing it.
- But it’s a bit rough on the people you send to prison, since what they did wasn’t exactly illegal when they did it.
You could certainly object to this theory. One very obvious objection would be: Come on, of course lying to customers about the price of a bond is fraud, and was always fraud, and so there is no problem with sending people who did it to prison. I would expect way more people to agree with that objection than with my theory.
Somewhat surprisingly, though, courts keep being sympathetic to my theory. We have talked about the Second Circuit’s decision reversing Jesse Litvak’s conviction, which doesn’t exactly rely on my theory, but which does have the nice effect of (1) prospectively making bond-price-lying a scary thing to do while (2) getting Litvak himself—who was the original poster boy for this crime and who was sent to prison for it—out of prison. We have also talked about a bunch of former Nomura Securities International Inc. RMBS traders who were charged with similar crimes; a jury ended up convicting one of them (Michael Gramins) and deadlocking on another (Ross Shapiro).
But yesterday the federal district judge in Connecticut overseeing that case threw out Gramins’s conviction and ordered a new trial, and while the reasoning there was similar to that in the Litvak case—basically, prosecutors put on evidence that Gramins’s customers trusted him not to lie to them, which they shouldn’t have—there is a lot of commentary to the effect of, come on, really, lying to customers was a crime? For instance (citations omitted):
I agree with the defendants that this case raises due process concerns with regard to both fair notice and discriminatory enforcement. As the Government concedes, lying in arms-length commercial transactions is not always illegal. It depends on the particular facts and circumstances. Prior to the indictment in Litvak, the conduct at issue appears to have been widespread in the RMBS market. Cooperating witnesses in this case testified that they didn’t realize the conduct was illegal. After a series of trials involving five defendants charged with essentially the very same conduct, only Mr. Gramins currently stands convicted on any count in any of the indictments. It is possible the jury convicted him on the conspiracy count, and hung or acquitted as to all other counts in the indictment, only because it was persuaded that he continued to engage in the conduct notwithstanding the Litvak indictment. Others who engaged in this conduct have been the subject of civil enforcement proceedings or no enforcement proceedings at all. It is fair for the defendants to wonder what distinguishes their conduct from that of others who have been spared indictment.
And while the judge disagreed with Gramins that the prosecutors’ statement that “lying to take people’s money is a crime” was so outrageous as to warrant a new trial (which he ordered for other reasons), he did agree that the statement was wrong:
More problematic in the context of this case is the prosecutor’s admonition to the jury that “lying to take people’s money” is a crime. The defense objected and a curative instruction was provided on the intent and willfulness elements of the charged offenses. Though the prosecutor’s repeated statement was at odds with the Court’s instructions on the law, it does not appear that his oversimplification of the law was calculated to inflame the jury. Moreover, it does not appear that the jury was misled. If the jury believed that “lying to take people’s money” is a crime, it likely would have convicted all the defendants on all counts.
It is intuitive, straightforward, and I would have thought popular to say: Look, whatever you think about the odd norms of the sophisticated mortgage-backed-securities market, lying to take people’s money is a crime, and so even if these traders thought it was okay, we can still send them to prison for doing it. And that is more or less what prosecutors said. But it is not quite what the law says.
Read the … source code?
Here is a pretty amazing story about an Australian crypto-exchange company (Byte Power Party Ltd.) that sold a stake in itself to a Singaporean crypto company (Soar Labs) in exchange, mostly, for Soar’s in-house crypto token, Soarcoin, which lives on the Ethereum blockchain. But then Soar “accused Byte Power Group of not selling its Soarcoin at ‘manageable levels’”—I leave it to you to interpret what that says about the market for Soarcoin—and so it … took back the coins. Because the Ethereum smart contract for Soarcoin let it:
[A Byte Power] representative did say that "the way in which the smart contracts were written allowed them [Soar Labs] to remove the coins, which the company itself wasn't aware of at the time until the coins were actually taken."
On Tuesday, ISMG contacted Nicholas Weaver, a researcher with the International Computer Science Institute and a lecturer at the University of California at Berkeley. Weaver has studied virtual currencies and their surrounding ecosystems since 2013.
While on the phone with ISMG, Weaver browsed Soarcoin's code. Within about two minutes, he found a zero-fee transaction function that can only be called by the owner of the Ethereum smart contract, which in this case would be Soar Labs.
"If I'm the account owner, I can call that function and transfer a balance from anybody to anybody," Weaver says. "It's best described as a backdoor hiding in plain site."
Weaver says that Soar Labs could do what it did to Byte Power Party to anyone else holding Soarcoin. "The code says the owner of the contract can rewrite the balances at will," he says.
Super! Soar’s response is pretty much, yep, what of it:
Soar Labs' CEO Seth Lim tells ISMG that the issue found by Weaver isn't hidden. The code is open for anyone to see, he says.
"It's about trust in the ecosystem," Lim says.
When asked why Soar Labs didn't inform Byte Power Party, Lim indicated the company should have looked at the code.
Ah! We talked only yesterday about the limits of “just read the documents.” You can’t really have a financial system where everyone has to read all the documents all the time; efficiency demands that people sometimes be able to rely on trust and custom and things working as expected. I wrote:
Investing without reading the documents is, in almost all relevant cases, perfectly fine. I myself own shares of a Vanguard Group S&P 500 index fund. I have never read the prospectus for that fund, and I’ve certainly never read the annual reports of the vast majority of the 500 companies it comprises. I take a sort of efficient-markets approach to the whole thing: If page 37 of the prospectus said “also we are going to steal your money,” then I’d probably have heard about it, and Vanguard probably wouldn’t have trillions of dollars under management.
Soar’s smart contract literally said that. And Soar didn’t feel the need to mention it to a counterparty whom it was paying in Soarcoins, because that counterparty could always read the code. There is a certain cold logical appeal to that approach, but you could never run a real financial system based on it. Think how exhausting it would be if people spent all their time trying to hide Easter eggs like this in their source code, or looking for Easter eggs like this in their counterparty’s code. You’d never get anything done; everyone’s energy would be spent on trickery and counter-trickery rather than economic activity.
Elsewhere in the crypto, here is a story about what happens when crypto-hucksterism meets conceptual-art hucksterism; it’s a lot of stamping blockchain addresses on paper with blood, that sort of thing. And Susquehanna International Group is already doing a lot of crypto trading.
Nobody quite knows what market manipulation is, but in general it is fairly clear that if you buy stock for the express and sole purpose of keeping the stock price up, that is probably manipulation. So nobody does that, or at least, if they are doing it, they say that they are doing something else. But there is an important exception, which is that when a company does an initial public offering of stock in the U.S., its underwriters will agree to “stabilize” the stock: If the stock looks like it is going to fall below the IPO price in the day or so after it starts trading, the underwriters will step in to buy stock to keep the price up. (This is also common in follow-on offerings as well as IPOs.) “The underwriters may bid for, and purchase, shares of Class A common stock in the open market to stabilize the price of the Class A common stock,” an IPO prospectus might say. “These activities may raise or maintain the market price of the Class A common stock above independent market levels or prevent or retard a decline in the market price of the Class A common stock.” They are manipulative trades intended solely to keep up the price of the stock, and the company and underwriters say as much. It is perfectly legal, but it is not legal because it is not manipulative. It is legal because it is legal. It is legal because the U.S. rules around public offerings allow for a certain amount of this sort of manipulation, because the regulators share the view of the underwriters that “stabilizing” the price of a newly-public stock is a good idea.
This is not just an exemption from stock-manipulation rules. The underwriters also team up and agree on how they will trade. (Generally only one bank does the stabilization, on behalf of the other banks.) Ordinarily it is … problematic … if big traders coordinate their trading of a company’s stock. But in IPO stabilization, it isn’t. Because the rules allow it. (Or, at least, because the courts have decided that the rules allow it.) It’s not that they’re not forming a group to manipulate the price of a stock. It’s just that they can.
Anyway this is weird:
Australian prosecutors charged the former local heads of Citigroup Inc. and Deutsche Bank AG with criminal cartel offenses, the most dramatic move so far in a country bent on taking bankers to task for years of misbehavior.
The case centers on the banks’ actions following a A$2.5 billion ($1.9 billion) share sale by Australia & New Zealand Banking Group Ltd. in 2015. … The case centers on ANZ Bank’s institutional placement of 80.8 million shares in August 2015 and how underwriters Citigroup and Deutsche Bank disposed of the roughly 25.5 million shares they had to soak up in the sale. The third underwriter, JPMorgan Chase & Co., hasn’t been charged and has declined to comment on media reports that it received immunity in return for information.
The Australian Financial Review reported Monday a recorded video conference call was expected to show ANZ and its investment banks negotiating how about A$789 million of shares which went unsold in the placement would be offloaded into the market in order to minimize any downside risk to ANZ’s share price.
In the purest sense, when a bunch of banks get on a phone call to discuss how to coordinate trading a company’s stock in order to keep the price up, that sounds bad. But when they do it in connection with an underwritten offering, then it is … maybe fine? Whether it is fine depends on what they do, and on what the rules say. The banks say the rules were unclear:
“This is a highly technical area and if the ACCC believes there are matters to address, these should be clarified by law or regulation or consultation,” Citigroup said in a statement.
That statement goes on:
Underwriting syndicates exist to provide the capacity to assume risk and to underwrite large capital raisings, and have operated successfully in Australia in this manner for decades. In the transaction in question, ANZ raised important equity capital through the underwriting arrangement with the syndicate of banks, including Citi. Citi and its employees acted with integrity and without any bad intent in fulfilling the obligations of this underwriting agreement. As required by the Market Integrity Rules, Citi also effectively participated in orderly capital markets to ensure that the required outcomes for ANZ and its shareholders were achieved.
I don’t know exactly what the banks are accused of doing, nor exactly what Australia’s rules say, so I have no idea how this will go. But in broad outlines you can see the problem. Banks, and capital markets generally, are used to a certain set of rules that apply in normal circumstances, and to another set of rules that apply specifically in underwritten capital raisings. People who don’t spend a lot of time thinking about underwritten capital raisings, and who then discover the norms that apply there, often find those norms disturbing. Every so often those people are regulators and it all gets very awkward.
Twitter Inc.’s stock will be added to the S&P 500 Index before the market opens on Thursday. When a stock joins the S&P 500, a lot of S&P 500 index funds have to buy it, so it tends to go up. When a lot of people are buying a stock, that is a good time to sell that stock, if you have any. In particular, if you are the company that issues the stock, and people are clamoring to buy your stock, why not sell some? And so selling stock into an index add is a pretty smart and sensible way for a company to raise money. Capital-markets bankers know this, of course, and when a company is about to be added to an index, you can find bankers floating around suggesting that it use their services to raise some money.
The simple way to do this is to just sell stock, but that is a bit of an ugly look. Many companies prefer not to sell stock if they can help it: It is dilutive, it suggests that they think their stock is fully valued, etc. Selling a convertible bond—which, if all goes right, is like selling stock at a premium to today’s price—is arguably better messaging. And so that’s what Twitter is doing, selling $1 billion of six-year convertibles. It is not directly selling stock, but the arbitrageurs who buy the convertibles will be selling stock to hedge them, and they will probably find enthusiastic buyers among the index funds.
Twitter is also doing a “call spread” along with the convertible, where it synthetically buys back the conversion option in the convertible bonds from the banks underwriting them (called a “convertible note hedge”) and sells those banks a similar option at a higher strike price (called a “warrant”). It did the same thing in 2014, when it issued $1.8 billion of convertibles with a conversion price of $77.64 per share—a premium of 47.5 percent over the $52.64 closing price of its stock at the time—and used a call spread to synthetically raise the conversion premium to 100 percent. The idea is that if you don’t like selling stock at your current stock price, but prefer selling stock at a 47.5 percent premium to your current stock price, you’ll like a 100 percent premium even more. (Meanwhile, the convertible-bond market doesn’t particularly want that up-100-percent convertible, but the banks will happily make you a synthetic one.) Of course you pay up for that higher premium: Twitter pays the banks some cash for the note hedge, and the banks pay Twitter some cash for the warrant, but the note hedge (a closer-to-the-money option) is worth more, and so Twitter pays more than it receives. But there are some neat tax advantages to the call spread: Basically, Twitter gets to deduct what it pays for the note hedge but doesn’t have to count the warrant proceeds as taxable income.
This is all quite standard stuff, and, disclosure, I used to do it for a living. (I mean I did other things too but this is the bread-and-butter work of a “corporate equity derivatives” desk like the one I used to work on.) There are dozens of deals like this every year, but every time a big company does one, people email and tweet at me saying “what on earth is this witchcraft.” Because to be fair the disclosure of it is not super-clear; the announcement is full of a lot of legal rigamarole about how the banks plan to hedge their options (basically: buy stock, probably from convertible arbitrageurs), and is conspicuously lacking, say, a clear diagram of what exactly is happening. The basic idea is simple—sell investors a bond with an option to convert into Twitter stock at a premium, buy that same option from the banks, and sell the banks a similar option at a higher premium—but the lawyers tend to make it sound more complicated.
Everything is seating charts, also bathrooms.
Man, here is a bit of symbolism about the respective value that WhatsApp and Facebook place on privacy:
Some employees even took issue with WhatsApp’s desks, which were a holdover from the Mountain View location and larger than the standard desks in the Facebook offices. WhatsApp also negotiated for nicer bathrooms, with doors that reach the floor.
At Facebook, the bathroom doors don’t reach the floor so that they can slide ads under them.
The desks-and-bathrooms warfare is from a story about what’s “Behind the Messy, Expensive Split Between Facebook and WhatsApp’s Founders,” which is actually pretty much just what you’d expect: WhatsApp’s founders thought that targeted advertising is bad, that privacy is good, and that ads are “insults to your intelligence,” but then they made the in-hindsight-obvious mistake of selling their service to Facebook for $22 billion. Facebook, meanwhile, thinks that targeted advertising is good, that privacy is bad, that your intelligence is nothing to write home about anyway, and that—crucially—it ought to make a return on its $22 billion investment. You can fill in most of the details from that basic conflict, though the bathroom stuff was a surprise.
Elsewhere in … tech … sure, let’s say tech … “Uber CEO Dara Khosrowshahi insists that leaders say they ‘have the D’ in meetings—and bewildered employees aren’t sure if he gets the other meaning.”
Hedge Fund Managed by Billionaire Howard Gained 36% in May. Capstone: the fund betting that market volatility has a future. Andreessen Horowitz, Sequoia Big Winners in GitHub’s Sale to Microsoft. As Oil Soars, Few Hedge Funds Are Left to Profit. Former Goldman banker Byron Trott plans fundraising. Why Banks Are Losing the Battle for M.B.A. Talent. Trump Unable To Remember Words To "God Bless America" At Performance Commissioned To Prove His Patriotism. A Russian Oligarch’s $500 Million Yacht Is in the Middle of Britain’s Costliest Divorce.
If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Thanks!
Like Money Stuff? Subscribe to Bloomberg All Access and get much, much more. You'll receive our unmatched global news coverage and two in-depth daily newsletters, The Bloomberg Open and The Bloomberg Close.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the editor responsible for this story:
James Greiff at firstname.lastname@example.org