Citi Spoofing and Yahoo Hacking

Matt Levine is a Bloomberg View columnist. 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 Third Circuit.
Read More.
a | A


It seems like pretty big news that Citigroup Global Markets Inc. agreed to pay a $25 million penalty to the Commodity Futures Trading Commission last week to settle charges that it spoofed the U.S. Treasury futures market. Like ... Citi is a big bank, and Treasury futures are a big market, and $25 million is a big fine. It's the biggest spoofing story since that time a random dude in London maybe crashed the stock market.

But the CFTC and Citi don't seem to be very interested in telling the story. The CFTC order is brief, bland and generic. Five Citigroup traders "engaged in the disruptive practice of 'spoofing' (bidding or offering with the intent to cancel the bid or offer before execution) in U.S. Treasury futures markets" in 2011 and 2012, and they did it more than 2,500 times. (Bloomberg News has their names.) The "spoofing strategy involved placing bids or offers of 1,000 lots or more" -- $100 million or more face amount -- "with the intent to cancel those orders before execution"; they put in the spoofing orders "after another smaller bid or offer was placed on the opposite side of the same or a correlated futures or cash market," and cancelled them "after either the smaller resting orders had been filled or the Traders believed that the spoofing orders were at too great a risk of being executed."

That is how one spoofs, yes. But there are no funny e-mails or chats, no details of the trades or profits, no sense of how or why this happened, except that it happened a lot. Well, there is one anecdote, about a junior trader in Tokyo who visited the New York desk for training. He was apparently trained in spoofing, but not very well:

In January 2012, the junior trader returned to Tokyo and began placing spoofing orders. In one instance, on January 31, 2012, he placed a 4,000 lot offer in the 10-year U.S. Treasury futures market that he intended to cancel before execution to induce other market participants to trade on his smaller resting bid in the 10-year cash U.S. Treasury market. The majority of his 4,000 lot spoofing order traded before he could cancel the order and, as a result, he incurred a loss. Afterwards, the junior trader called several other members of the U.S. Treasury desk, including the head of the desk, to report his loss and explain his spoofing strategy and what had happened. Although the supervisor told the junior trader "that's not a smart thing to do" and cautioned him not to do it again, neither the supervisor nor the other members of the U.S. Treasury desk reported the incident to compliance or any other senior manager, despite having information that the junior trader engaged in spoofing.

Did the supervisor mean that spoofing is "not a smart thing to do," or just getting executed on your spoof order? Because the latter is inarguable.

The former is a little inarguable too? People talk about "white-collar crime" (fraud, etc.) as distinct from "blue-collar crime" (burglary, etc.), but there is a social hierarchy of fraud, too. Big banks might collude on a benchmark fixing, or sell mortgage bonds with misrepresentations about the level of due diligence they conducted, but they don't, like, run Ponzi schemes. Individual junior bankers might insider trade here and there, but big banks' trading desks don't make their money by insider trading. This is not because banks are paragons of morality; it's because they are big institutions with big compliance departments who have read the rules and thought about ways to prevent well-known sorts of dumb obvious misbehavior. Perhaps this leads to a culture of compliance, or perhaps it leads to a culture of baroque sneaky misbehavior, but in any case it probably does catch the low-hanging fruit. 

These days spoofing feels pretty blue-collar, and most spoofing enforcement cases so far have tended to be against small firms and independent traders. And Citi's spoofing -- putting in 1,000-contract orders, trading small orders on the other side, and canceling the 1,000-contract orders, over and over again, thousands of times -- just feels too big and dumb and obvious for a big bank. But here we are. I suppose things were different in 2011 and 2012, when regulators were just beginning to take spoofing seriously. Sometimes it takes time for misbehavior to become dumb and obvious.


Here is a story about how the Securities and Exchange Commission is investigating Yahoo Inc. because it was hacked in 2014 and waited two years to disclose the hack to shareholders. (And users! It waited two years to disclose to users that they had been hacked! But that's not the SEC's concern.) On the one hand, sure, absolutely: You should disclose material stuff, and if you don't, then investors are misled into buying your stock at inflated prices, and then they are harmed when the stuff is disclosed and the price drops.

The weird thing about Yahoo, though, is that it waited so long to disclose the hack that, by the time it did, it had already agreed to sell itself to Verizon Communications Inc. (Well, to sell the Yahoo bits of itself. It's keeping some other bits, and renaming the stub "Altaba.") And Verizon hadn't known about the hack either. Now it does, of course, though it is still unclear whether it will try to re-negotiate the deal because of the hacks.

But this makes the SEC's potential case weird, especially if the deal closes as scheduled for the agreed price. Like: If Yahoo agreed to sell its business for a fixed amount of cash before disclosing the hack, and then it sells it for that amount of cash after disclosing the hack, then was the hack material? Shareholders got the same amount of money from Verizon after the disclosure as they expected to before the disclosure. And if the SEC ends up fining Yahoo for not disclosing it, who pays the fine? Verizon is getting the guilty businesses, but it seems a bit rough to make Verizon pay more for a hacked Yahoo than it had originally agreed to pay for an un-hacked Yahoo. Presumably Altaba would pay, which would mean that Yahoo's disclosure breach deceived shareholders but didn't cost them any money -- but that the SEC's efforts to protect those shareholders would cost them money.

Active ETFs.

One problem with mutual funds is that if you buy a mutual fund, and don't sell it, you still have to pay capital gains taxes on it. This is not true of most other investments: If you buy stock, and it goes up, you don't pay taxes on the appreciation until you sell. Certainly if the stock doesn't go up, and you don't sell, you don't pay taxes. But "a mutual fund manager must constantly re-balance the fund by selling securities to accommodate shareholder redemptions," which "creates capital gains for the shareholders, even for shareholders who may have an unrealized loss on the overall mutual fund investment."

A big advantage of exchange-traded funds is that they don't do that: When you want out of your ETF, you sell your shares to another investor, or get redeemed in kind, so the ETF is not constantly selling securities (and incurring taxes) to meet redemptions.

The ETF approach feels right, and the mutual-fund approach feels dumb, which is why people who manage mutual funds are constantly trying to turn them into ETFs. If you run an index fund where everyone always knows what's in it, this is pretty easy. If you run an actively managed fund, it's hard. The point of an ETF is that it is exchange-traded: People know what securities the ETF holds, they know what those securities are worth, and they can make sure the ETF's price is right by arbitraging between the ETF and the underlying securities. An active fund whose holdings are secret, or semi-secret -- active funds disclose their holdings periodically, but not minute-by-minute -- can't be effectively arbitraged.

But of course the active managers don't care about that. They just want the tax advantages of ETFs, and will take convoluted approaches to get them. So here is a story about how J.P. Morgan Chase & Co. "is joining a long line of asset managers" hoping to license Precidian Investment's "ActiveShares" nontransparent ETF model, which "would disclose stockholdings on a delay rather than daily," if the SEC approves it.

I am a little torn. On the one hand, the ETF model -- of ensuring price accuracy by arbitrage -- is an elegant one, and the nontransparent ETF model -- of ensuring price accuracy by, um, "Precidian would post the real-time value of its holdings on exchanges" -- feels kludgy and disappointing. On the other hand, the taxation of regular mutual funds really is pretty dumb. What the active managers really want is just an incantation that allows them to run a regular mutual fund, but to be taxed like an ETF. I am sympathetic. In an ideal world, that incantation would just be checking a box that says "I'd like to be taxed like an ETF," rather than having to do weird stuff to look kind of like an ETF.

By the way, the reason people want active ETFs to be nontransparent is to avoid "giving away their proprietary trading strategies to potential copycats," and I always wonder a little if that concern is overblown. Is the worry that someone will "front-run," as it were ("back-run"?) a big fund by seeing what positions it is buying into (over several days) and racing to buy ahead of it? Or is it that, if you have a day-by-day breakdown of how a big famous expensive fund invests, you could build your own generic copycat fund and market it to investors cheaper? "The Schmimco Schmotal Schmeturn Fund follows the same strategy as the leading brand, but at half the management fee!" That would be a fun business, until it's eaten by robots.

Snap votes.

We talked last week about Snap Inc.'s proposed initial public offering, which would give public shareholders non-voting stock and no control over how the company is run. I said that journalists and academics and governance experts love to talk about how bad dual-class share structures are, but then big investors buy into dual-class IPOs anyway, and with no evident price discount. Founders are never really forced to choose between maintaining control and maximizing their economics. So of course they keep doing dual-class IPOs.

But then this article came out about how "T. Rowe Price is pushing back on Snap's plans to only sell non-voting shares in its IPO," and I thought, well, that doesn't mean much. Just because T. Rowe says it is "quietly and persistently" objecting to the dual-class IPO doesn't mean that it will actually pay less for new Snap shares. (T. Rowe is already an investor in Snap via an earlier private investment round.) But it's actually even less meaningful than that: T. Rowe actually went and issued a press release saying that "contrary to certain news reports, T. Rowe Price is not contesting the plan by Snap Inc. to issue non-voting shares in its impending initial public offering." They "generally do not favor proposals that would create disproportionate voting rights," but, you know, for Snap, sure.

If you generally don't like dual-class stock, but you can live with each specific instance of dual-class stock, you will get a lot of dual-class stock. The right way to oppose dual-class stock is to oppose it even in companies you like. But no one is willing to go quite that far.

Elsewhere, "Snapchat Is Justifying Its $20 Billion Valuation by Emphasizing User Engagement." And Snapchat seems to be cracking down on clickbait?

On Snapchat last week, MTV published an article with the headline “Is this the thirstiest person on earth?” The article, which ran on Snapchat’s news service, Discover, appeared with a picture of a bikini-clad blond woman taking a selfie, even though the piece was about a fully clothed man.

Those kinds of risqué and misleading images will now be much less prominent on Discover because early on Monday, Snapchat updated its guidelines for publishers in a way that essentially cleans up what is shown on its news service.

That is the state of the media industry in 2017: A messaging service designed to allow college students to share nudes is now exercising its editorial judgment to tell major media companies to maybe cut it out with the misleading bikini pictures.


Here's a story about insurance. Basically you buy an insurance policy from Insurer X, and then you submit claims, and Insurer X pays the claims. Or it doesn't -- "all insurers have reasons to minimise claims costs" -- but if it turns you down, you have leverage. Not only can you sue to hold Insurer X to the contract, but you also have a business relationship: If Insurer X wants to sell you another policy, it had better pay your claims on this one. But Insurer X can extract some value by ceding your policy to Reinsurer Y, which has never met you, will never sell you insurance directly, and can feel free to stonewall your claims as far as it's allowed by law. The particular value extraction here comes from not being nice: By transforming the insurer/insured relationship from a customer-service relationship to a pure standoffish contractual relationship, the insurer can make it a bit more profitable.

The fun part of this story is that Insurer X is American International Group, and Reinsurer Y is Warren Buffett's Berkshire Hathaway Inc., and it is funny to think that AIG is too nice and so is calling in Warren Buffett so he can be mean.

Oh, you know, the stuff.

Davos is all-in on Secret Trump Theory

Executives gathered in the Swiss resort for the World Economic Forum this week keep repeating, like a soothing mantra, that Donald Trump is at heart a pragmatist who will avoid trade wars and regulations that make it harder to do business.

“What somebody’s saying is not necessarily what they’re going to do,” said David Cote, chief executive officer of Honeywell International Inc.

Hey great super. Elsewhere, here's a doomed lawsuit over Trump's likely Emoluments Clause violations. Here's some advice for federal employees. And: "Habitual presidential stock tweeting should increase the component of total portfolio risk associated with individual stock risk." And: "Trump Aide Says Press Secretary Used ‘Alternative Facts.’" And: "China Slams Western Democracy as Flawed." And: "All populists are at heart conspiracy theorists, who pretend that easy solutions exist to society’s woes and have only not been tried to date because elites are wicked and deaf to the sturdy common-sense of decent, ordinary folk." And: Trump "gets bored and likes to watch TV." And: Trump apparently plagiarized Obama's inaugural cake

People are worried about stock buybacks.

One weird thing about stock-buyback worrying is its asymmetry. If you are worried that companies are buying back too many shares, then you should be equally worried that they aren't issuing enough shares. There's nothing magical about the number of shares a company has now. If it's bad for a company to spend $100 million on buybacks instead of on building new factories, why stop there? Why shouldn't the company sell $100 million of stock, and then spend $200 million on factories? Or $300 million? Why is the right amount to spend on factories always the amount of money that the company has now? Facebook recently announced a $6 billion stock buyback, but imagine if Facebook had instead announced a $6 billion stock offering. That would have been much weirder, right?

Here is Bloomberg Gadfly's Lisa Abramowicz on debt-financed buybacks:

This is basically taking money from debt buyers to pad the pockets of equity holders. It has no obvious benefit to the company's long-term prospects; in fact, quite the opposite. As Fitch Ratings said in a report on Thursday, "leveraged share buybacks and other shareholder friendly actions are an ongoing risk to bondholders as borrowing costs remain historically low."

But is the concern here that companies have too much debt and too little equity, or is it just that they have more debt and less equity than they did before the buybacks? If you are a company with $200 of stock and $100 of debt, and you issue $50 of new debt to buy back $50 of stock, then you will have more debt and less stock. But how did we know that $200 of stock and $100 of debt were the right amounts of each? What if that was too conservative, and you were just adjusting your capital structure to get it back to a more normal ratio?

That is to some extent an unanswerable aesthetic question; more realistically, it is a question that can only be answered through the economic cycle. (You'll know, eventually, if you had too much debt!) But it is not particularly obvious that big U.S. public companies have too much debt and too little equity, in historical context:

Source: Bloomberg

I suppose that chart cheats by including financials, but still. Today, S&P 500 companies have debt equal to about 112 percent of their equity. Ten years ago, that number was 213 percent. If you didn't think then that they should issue stock equal to half of their market capitalizations to pay down debt, why would you think now that they shouldn't issue debt to buy back stock?

Me Friday.

I wrote about Wall Street research and "corporate access." It's worth pointing out that the U.K. has gotten rid of this issue: The Financial Conduct Authority banned the use of research payments for corporate access back in 2014. It still seems to be a major driver of research payments in the U.S. though.

Things happen.

Planet Money on Ed Thorp. Odd Lots on Islamic State monetary policy. Myanmar's Yangon Stock Exchange is pretty boring. Central Banks Embrace Risk in Era of Low Rates. Forbes settles long-running takeover dispute. Struggling hedge funds still expense bonuses, bar tabs. Here’s Another Way Wells Fargo Took Advantage Of Customers. Quicken Loans, the New Mortgage Machine. How Well Does Running Vanguard Pay? Abu Dhabi Merges State Funds to Form $125 Billion Investment Firm. The tip of the iceberg: the implications of climate change on financial markets. Fannie Mae Shareholders Lose Another Round in Federal Court. The carp industry. Kristen Stewart has co-authored a paper on artificial intelligence. Are There Any Banks on Bank Street? "XL Specialty Insurance Co. asked a New York federal judge Thursday to order a Manhattan art gallery to reimburse the insurer for covering a painting the owner reported stolen when it turns out he merely forgot he had sent it to be reframed." Tinder for adoption. This Crazy Fan Theory About ‘Jeopardy!’ Actually Makes Total Sense. "And there are quite a number of birds that fly." 

If you'd like to get Money Stuff in handy e-mail form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at

To contact the editor responsible for this story:
James Greiff at