Snap Votes and Banker Bonuses

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

Disappearing votes.

There is a bit of a tradition of initial public offerings by hot internet companies where the founders sell stock to the public but keep control for themselves by setting up dual-class share structures and giving themselves high-voting shares. But then later they realize that they haven't given themselves enough votes -- if they issue new stock for acquisitions, or sell some of their stock for philanthropic purposes, then they'll lose control of their companies --- and so they have to sort of shamefacedly creep back to the shareholders and say "hey would you mind giving us even more votes?" It's fine, really -- the founders control the company, so what are the shareholders going to do, say no? -- but it's a little awkward. People complain. You end up with triple-class stock. It all looks a little dumb.

There is absolutely no need to do it like that. Founders could just give themselves total permanent control at the time of the IPO: Instead of giving themselves 10-vote shares and the public 1-vote shares, or whatever, they can give themselves 10-vote shares and the public zero-vote shares. Then they never need to go back to the public shareholders to ask them to approve a new share class, or anything else. The founders can just run the company however they want, and public shareholders will be limited to putting up cash and hoping for the best.

Snap Inc. is apparently planning to do it the right way:

Investors won’t get any voting power with shares purchased in Snap’s initial public offering, according to people familiar with the matter.

That leaves key decisions, such as the makeup of the board, primarily to Evan Spiegel and Bobby Murphy, co-founders of Snap, the owner of the disappearing-message app Snapchat. The two are expected to hold more than 70% of the voting power despite owning roughly 45% of the stock, the people said.

I appreciate their tidy-mindedness, which is not particularly evident in their app's user interface. (The other 30 percent of the voting power will apparently go to other pre-IPO investors.) 

There is the usual guff about shareholder rights ("Some big investors contend that dual-class structures unfairly remove public shareholders from the decision-making process"), and about Snap's ability to impose this structure despite that pushback ("If you’re the only supply in the market, you’re well positioned to dictate the terms"), all of which feels a bit unreal. No one cares about this stuff. I mean, people care: Journalists and academics and governance experts love to talk about how bad dual-class share structures are. But in financial markets, there is only one meaningful way to care, and that is in price. If IPOs with dual-class shares came at a discount to IPOs with full shareholder rights, then this would matter. Bankers would go to founders and say: Look, you can keep control of your company by selling zero-vote shares to the public, but you'll raise less money than you would by selling full-vote shares. And the founders would have to make an economic decision about what they value -- informed by the fact that investors made an economic decision about what they value. But these are clearly not conversations that founders are having, and there's not much evidence of a price discount for dual-class IPOs. ("The difference in post-IPO performance between dual-class companies and non-dual-class companies isn’t statistically significant, both among tech companies and all U.S. listings.") Some people might care about dual-class stock, but the people actually buying the stock don't care enough to pay less for it, meaning that in a practical sense no one cares.

One way to think about this is: Look, there is nothing particularly magical about voting stock. Companies raise money in lots of different ways, and those ways give investors different rights. Bond investors usually don't get to vote on directors or mergers. Stock investors usually do. Snap stock investors ... won't. Voting for directors is one way that stock investors traditionally protect their investments, but there are other ways to protect your investment, and Snap investors might be reasonably confident that they'll get a good return even without that traditional protection.

Another way to think about it, though, is that no one cares because no one has ever had any reason to care. (Or at least not recently, and memories are short.) The high-profile Internet companies with the most egregious share structures -- Alphabet Inc. and Facebook Inc., which both went public as dual-class companies and then added a third class to cement founder control -- have made massive piles of money for their shareholders and are still run by active visionary founders who are liked by their investors. Eventually some tech founder will do some awful self-dealing, and investors will complain, but he'll get away with it because his dual-class stock allows him to ignore investors, so they will take their pique out on the next tech company that tries to go public with dual-class stock. (This happens in the bond market every now and then.) On the other hand, though, I don't know; people complained a lot when Elon Musk wanted to merge Tesla Motors Inc. and SolarCity Corp., and he was able to do it anyway, and they didn't have dual-class stock. Visionary founders can get away with a lot, whatever the corporate-governance formalities say.

Elsewhere: The corporate governance business is dominated by women. ("But when we are hiring, we need to really push that diversity to make sure we have men on the slate.")

Disappearing bonuses.

During and after the financial crisis of 2007-2008, bankers' bonuses became a focus of public attention, and there was a lot of very heated and very boring debate about them. It went something like this:

Regular people: In my business, a bonus is something I occasionally get for doing an extra-good job. Why are all these bankers getting huge bonuses when they obviously did a terrible job this year?
Bankers: In my business, a "bonus" is a misleading name for a portion of my expected compensation that is paid at the end of the year, and while its amount may vary depending on how good a job I had, it is in no way optional: It's paid every year, I rely on it, and in fact in most years it makes up the large majority of my pay.
Regular people: That all sounds pretty fake.
Bankers: Still.

Do you remember when the U.S. Congress almost imposed a 90 percent retroactive tax on bankers' bonuses? I do! (It was of some personal interest to me, at the time.) It is easy enough to understand that a money-losing company would still need to pay its employees' salaries, but people just could not get their heads around the idea that a money-losing company might also pay bonuses. The 90 percent tax thing never happened, but the European Union really did cap bonuses at two times base salaries just because it felt like bonuses shouldn't be that big.

Anyway time and economics seem to have resolved the misunderstanding:

Deutsche Bank, the former Wall Street powerhouse, may hold back on giving out bonuses to as many as 90 percent of bankers and traders, The Post has learned.

Only the top 10 percent of revenue generators may get a bonus for 2016 — and even then it will be paid out over the next five years, according to a source briefed on internal discussions.

If you only give a bonus to the top 10 percent of your employees, then it is a bonus all right! That fits pretty well with the popular understanding. Next year maybe Deutsche will pay its bonuses in the form of $100 Starbucks gift cards, and the popular and bank conceptions of "bonus" will converge entirely, at least at Deutsche Bank.

I have always been pretty skeptical about restrictions on banker bonuses. They seem to rest on arithmetic errors; paying someone a $1 million base salary and a $1 million bonus is in many ways riskier for the bank than paying her a $200,000 base and a $1.8 million bonus. But if banking bonuses become exceptional rewards for a job well done, and if most bank employees expect to get them rarely or never, then that really will change banks' culture. People talk about making banking boring again, and taking away bonuses will make it very boring indeed. If that's what you want!

Meanwhile, in any other context you'd probably be annoyed by this:

Deutsche Bank AG has banned text messages and communication apps such as WhatsApp on company-issued phones in an effort to improve compliance standards.

But now it's like:

Manager: Hey, just FYI, you can't text on your company phone anymore, and also you're not getting a bonus.
Banker: But I love tex-- wait, what?

Bitcoin HFT!

I assume the Venn diagram of "people who invest in bitcoin" and "people who think that the financial system is rigged against them because high-frequency traders are allowed to co-locate their servers and front-run ordinary investors" has a lot of overlap, so what about this, huh?

The reality is that professionals armed with cutting-edge technology now drive as much as 80 percent of bitcoin trading, mimicking strategies honed by some of the biggest players on Wall Street. To them, bitcoin is just the latest asset class ripe for conquering with machines.

The cryptocurrency’s market structure ticks all the right boxes: arbitrage opportunities across multiple exchanges, zero transaction costs on Chinese venues that host most of the world’s turnover, round-the-clock trading, and co-location services allowing participants to place their servers right next to those of the exchange.

Also I am going to go out on a limb here and guess that the risk of adverse selection in the bitcoin market is somewhat lower than it is in, say, the U.S. stock market. You're not trading against Fidelity here, you know? The basic raison d'être of high-frequency trading is that it makes markets more efficient, and that's not just a matter of enforcing the law of one price by arbitraging bitcoin prices across multiple platforms. It's also about efficiently extracting value from amateur noise traders and giving it to professionals. 

Pound flash crash.

Here is a Bank for International Settlements report on "The sterling 'flash event' of 7 October 2016." In the BIS's telling, it was the standard feedback loop of any market crash: Some people wanted to sell pounds, so the price of the pound declined, which led to further selling to hedge option positions, so the price declined some more, and stop-loss orders were triggered, leading to even more selling. "The presence, outside the currency’s core time zone, of staff less experienced in trading sterling, with lower risk limits and risk appetite, and with less expertise in the suitability of particular algorithms for the prevailing market conditions, appears to have further amplified the movement." Also:

This event does not represent a new phenomenon but rather a new data point in what appears to be a series of flash events occurring in a broader range of fast, electronic markets than was previously the case in the post-crisis era, including those markets whose size and liquidity used to provide some protection against such events.

At the same time it's not as simple as saying "high-frequency electronic trading causes instability":

Some firms withdrew first from voice trading, others from e-trading. Such halts lasted anywhere between two and 30 minutes. Some recommenced trading automatically when conditions stabilised, but many required management override. For some it was their algorithmic trading that restarted first (although often with human intervention to allow the restart), for others their voice activity. Those withdrawing liquidity cannot be readily categorised by type of institution – various major dealers, principal trading firms and firms representing a retail client base confirmed that they withdrew liquidity provision from the market during the event.

The story is that the price dropped, so some traders stopped trading, which made liquidity worse, and then later they started trading again, which made it better. Some of that trading was electronic, and some of it wasn't, but there's no tight connection between electronic trading and withdrawal of liquidity.


One basic problem with regulation is that you probably want some good thing, so you make a rule saying "there shall be the good thing," but the rule doesn't just magically create the good thing. Instead, it creates an enforcement mechanism. The enforcement mechanism may tend to create more of the good thing, but enforcement mechanisms in themselves are sort of necessarily bad. They're coercive, they're punitive, they're bureaucratic, they're inflexible, etc. You can't create rules just because you want the good thing; you have to also want the enforcement mechanism. If you forget this, you get terrible consequences. So you don't like drug addiction, so you say "there shall be no drugs," and you get mass incarceration. (And still drug addiction!) 

Anyway here is Cliff Asness on the Labor Department's fiduciary rule:

In what instances might we now hold an advisor liable for “not acting in a client’s best interest” but wouldn’t have under the old suitability standards? More important to my argument, in what ways might upping and broadening the standard instead create other problems for investors? I can think of four prominent possibilities. First, judging investments too much by ex post performance. Second, not judging investments in a portfolio context. Third, over-emphasizing low fees as always being in the client’s best interest. Fourth, judging innovative investing approaches more harshly than conventional ones. 

The point is that it would definitely be good if financial advisers acted in their clients' best interests, but you can't just decree that that will happen. All you can do is create an enforcement mechanism -- regulatory penalties and/or private lawsuits for advisers who don't act in their clients' best interests -- and that mechanism will not correspond precisely to the good thing that you want.

What is Bernie Madoff up to?

This is so perfect that I can't quite believe it:

“Bernie really was a successful businessman with quite original insights into the market, and he’s continued applying his business instincts in prison,” Fishman said. “At one point, he cornered the hot chocolate market. He bought up every package of Swiss Miss from the commissary and sold it for a profit in the prison yard. He monopolized hot chocolate! He made it so that, if you wanted any, you had to go through Bernie.”

I like that Bernie Madoff is enjoying prison so much because it allows him to be his true self. All those years, he was a massive crook, but had to go around pretending to be an upstanding businessman. Now, behind bars, he is finally free to let his crook flag fly. 

People are worried about unicorns.

As far as I can tell the list of top worries about unicorns goes something like this:

  1. Is there a bubble in which venture-backed tech companies are overvalued and will never be able to maintain their value when they go to public markets? 
  2. Does Silicon Valley's utopianism mask a reality in which the tech industry mostly solves the trivial problems of the rich by immiserating the poor?
  3. What if the unicorns are actually vampires?

"For $8,000 this startup will fill your veins with the blood of young people," says this article, and this is not the first time we have talked about vampirism in the tech world. In addition to being an excellent plot for a television thriller, it is a lovely parable of Silicon Valley. The blood startup is called Ambrosia, and while it has some vague basis in medical research (it "was inspired by studies on mice that researchers had sewn together, with their veins conjoined, in a procedure called parabiosis"), its clear inspiration is not science but myth, an atavistic throwback to magical notions of preserving eternal youth through vampirism. Nerd culture is about fantasy and science fiction, not science, which is why unicorns are called unicorns. And remember what Marx said.

Elsewhere, Peter Thiel might run for governor of California in 2018, presumably on an "I will drink your blood, you delicious young bags of meat" platform. And Silicon Valley is moving to the right politically.

People are worried about bond market liquidity.

Well you know the problem:

With banks’ having less capacity to house securities on their balance sheets, off-the-run Treasuries have become more difficult to trade. That has fueled an expansion of a premium for the most liquid on-the-run notes.

And here is the solution:

Citadel Securities’s priority in fixed-income market-making this year will be to increase its presence in harder-to-trade and non-benchmark Treasury securities.

I mean, it's solution. Citadel is sort of the plug in solving problems about bond trading: If market-making is an economically useful function, and if banks should not be in the market-making business, then Citadel should probably do more market-making. Or someone else -- Jefferies? -- but if banks are really being forced out of the bond trading business, then someone has to pick up the slack, and Citadel always seems to float around these stories.

Elsewhere, here is a report from the Financial Stability Board on "Policy Recommendations to Address Structural Vulnerabilities from Asset Management Activities." The first structural vulnerability is of course "liquidity mismatch between fund investments and redemption terms and conditions for open-ended fund units," which you and I know is a bond-market-liquidity worry. The recommendations include more disclosure, "liquidity risk management tools" like "swing pricing, redemption fees and other anti-dilution methods," and stress testing.

Things happen.

May Ready to Announce Britain Will Leave EU Single Market. BAT Wins Reynolds With Sweetened $49.4 Billion Buyout Bid. JP Morgan to join Goldman in opening up risk system. JPMorgan Wants to Prove You’re Getting the Best Currency Trade. Rolls-Royce to Pay $817 Million to Resolve Bribery and Graft Inquiries. Moody’s Reaches $864 Million Subprime Ratings Settlement. Pursuit of Samsung Heir Over Bribery Threatens to Destabilize Company. How a Financial Council Republicans Loathe Could Work in Their Favor. Donald Trump Warns on House Republican Tax Plan. "Rep. Tom Price last year purchased shares in a medical device manufacturer days before introducing legislation that would have directly benefited the company." The Future of UK Debt Restructuring. "Bare your swords towards false western ideals like judicial independence." "Worry if no one thinks you are crazy. Supplement this with actually being crazy." Johnny Depp Sues Business Managers, Accuses Them of Fraud. Swiss Watchmaker Adds Cheese to ‘Make Swiss-Made Great Again’. Martin Shkreli Pelted With Poop At UC Davis As Yiannopoulos Event Gets Canceled Amid Protest. "One notable event is a simulation of a refugee’s experience, where Davos attendees crawl on their hands and knees and pretend to flee from advancing armies."

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