Header Ads

Money Stuff: There’s a New Volcker Rule Now

Money Stuff

BloombergOpinion

Money Stuff

Matt Levine

Volcker 2.0

In some broad sense, the only thing that a bank could ever do is "proprietary trading." The bank has some money—from shareholders, from depositors, etc.—and it uses that money to buy stuff. The stuff all comes with some risk,[1] so in some sense everything a bank does is making risky bets with its depositors' money. The old-timey bank that takes deposits and looks borrowers in the eye before giving them mortgages is making a bet, with its depositors' money, that the borrowers will pay back their mortgages. 

But in the aftermath of the 2008 financial crisis there was a lot of political desire to ban proprietary trading, and this desire had a strong you-know-it-when-you-see-it flavor. Obviously making mortgage loans is not proprietary trading; it's banking.[2] Something else is proprietary trading. But what? Clearly taking a bunch of customer deposits and using them to buy risky tranches of mortgage-backed securities in the hope that they'd go up and earn you a big bonus is proprietary trading, yes. But what about buying and selling loans in the secondary market? Parking excess deposits in Treasury bonds? In corporate bonds? Boring regular deposits-and-loans banks have lots of exposure to interest rates and credit; if they use derivatives to hedge those exposures, is that bad proprietary trading or good risk management?

The hottest controversies were over "market making," in which a bank stands ready to buy or sell, for its own account, securities and derivatives that its customers want to sell or buy. This is obviously "proprietary trading" in some sense—the bank is making bets with its own money in trading markets—and is also not a traditional banking activity; indeed, for decades, U.S. banks were largely prohibited from doing it, and it was left to investment banks. On the other hand market making is obviously not "proprietary trading" in the sense in which that term is normally used in the financial industry; market makers make a living by providing a service to investors and getting paid fees or spreads for doing so, while proprietary traders classically make their living by making correct bets on the future price moves of securities. (Loosely, market making is doing trades that your customers want to do; prop trading is doing trades that you want to do.[3]) And the fact that banks traditionally didn't trade securities, and investment banks did, was mostly a historical curiosity by 2008, especially after all of the big investment banks either became or were acquired by banks.[4] 

In 2010 Congress passed the Dodd-Frank Act, which included a provision calling for a "Volcker Rule" that would ban proprietary trading. U.S. bank regulatory agencies then spent years figuring out what that would mean. In 2013 they put out a final Volcker Rule, and it was honestly pretty good. It is not easy to agree on sensible intuitions about what constitutes "proprietary trading," and it is even harder to reduce those intuitions to a set of rules, and the regulators tried to, and the rule was very long, but it did a decent job of capturing stuff that felt like proprietary trading and allowing stuff that felt like banking, or lending, or market making, or underwriting, or hedging, or merchant banking, or other bank-y activities that are not "proprietary trading." It was overinclusive in some ways and underinclusive in other ways, and everyone had complaints. One particular complaint that big banks had is that the rule allowed them to continue to do market making, but added a lot of administrative hoops that they had to jump through to demonstrate that they were doing good market making and not bad proprietary trading. It is widely believed (and there is good evidence) that this actually cut back on banks' market making activities, which was bad for market liquidity.

This week the regulatory agencies put out a new and, broadly, less strict Volcker Rule. The general sense that one gets from reading news articles about the new rule is that it tweaks the Volcker Rule in ways that (1) help banks, (2) were lobbied for by banks, and (3) are too boring to discuss in detail. Broadly speaking if you are a little bank you are now exempt from more bits of the Volcker Rule than you used to be, and if you are a big bank your administrative burden in demonstrating compliance with the Volcker Rule is considerably reduced, and all of this may or may not cause or exacerbate some future financial crisis, but all in fairly boring and technical ways. Here are some good summaries of the changes from Bloomberg News. 

But you might browse through the (sure, fine, 301-page) release adopting the new Volcker Rule anyway, which is not just a discussion of technical changes but also a series of philosophical questions about what it is that banks do and are supposed to do. For instance: swaps. Sometimes a company takes out a floating-rate loan from a bank, but wants to fix its interest rate, so it asks the bank to enter into a swap agreement in which the bank will pay the company the floating rate and the company will pay the company a fixed rate. The loan is obviously a bank-y thing and not proprietary trading. But what about the swap? That is a derivatives trade made with the bank's own money. Is it a prop trade? 

The answer for big banks is quite straightforward: They are all market makers in swaps, they all hold themselves out as ready to buy or sell swaps for anyone, so if they do a swap with a corporate client it is part of their market making business and allowed under the Volcker Rule. But small banks don't make markets in swaps for hedge funds; they just go around making loans, and occasionally the client will want a swap with their loan, and the bank will want to provide it. Should it be allowed to?

Oh I don't know. Maybe, sure; after all, it is a client-service trade rather than an outright bet on interest rates. But also quite possibly not, right? Like, maybe little banks shouldn't be trading derivatives. (It is customary to think that these loan-related interest rate swaps aren't particularly risky, and in fact small banks generally hedge their risk by buying offsetting swaps from bigger banks. But several clients have notoriously blown up on loan-related swaps, with bad reputational effects for banks.) Or maybe, like … maybe if a client wants to take out a loan with a fixed interest rate, the bank should just give it a fixed-rate loan? The fact that the client gets a floating-rate loan with a swap, instead of the fixed-rate loan it really wants, is sort of an artifact of financialization, of securitization and syndication and risk-slicing: Basically investors want to buy loans with floating rates, and banks (even small ones) want to make loans that investors will buy, and so it is customary to split the interest-rate risk off from the credit risk in the form of a separate swap and loan. But if you're the sort of person who doesn't like proprietary trading, there's a decent chance you won't like any of that either.

But the small banks went to the regulators and said "we would like to do swaps," and the regulators couldn't really say no, because you can't really have a Volcker Rule that says "too-big-to-fail universal banks can do derivatives for their clients but small local banks can't," come on. So the new Volcker Rule says that big market-making banks can still do loan-related swaps as part of their market making business, and small banks can do them just, you know, anyway:

After considering the comments received, the agencies are excluding from the definition of "proprietary trading" entering into a customer-driven swap or a customer-driven security-based swap and a matched swap or security-based swap[5] if: (i) the transactions are entered into contemporaneously; (ii) the banking entity retains no more than minimal price risk; and (iii) the banking entity is not a registered dealer, swap dealer, or security-based swap dealer. … 

The agencies believe that adopting this exclusion will reduce costs for non-dealer banking entities and avoid disrupting a common and traditional banking service provided to small and medium-sized businesses. This exclusion will provide a greater degree of certainty that these customer-driven matched swap transactions are outside the scope of the final rule. …

The exclusion applies only to banking entities that are not registered dealers, swap dealers, or security-based swap dealers. This approach is consistent with feedback from commenters noting that primarily smaller banking entities have faced compliance challenges with respect to customer-driven swaps activities. Banking entities that are registered dealers, swap dealers, or security-based swap dealers generally engage in these activities on a more regular basis and therefore have been able to conduct their derivatives activities pursuant to the exemption for market making-related activities. 

This seems like a technical fix but it is not really driven by any sort of technical consideration; the question is really more, is this a banking thing or not? Is this a service that we want regular banks to be able to provide to customers as a matter of course? Is it "a common and traditional banking service"? I am not sure that that is really a question that can be answered from first principles, and I am fairly sure that, if you just asked a bunch of ordinary people and journalists and politicians if small banks should be able to do interest-rate swaps with their borrower clients, plenty of them would say no. But if you asked a bunch of bankers and bank regulators, I suspect they would almost unanimously say yes.[6] Because it is a service that banks have gotten used to providing, and that customers like to be able to get from their banks, and the Volcker Rule authorities try to err on the side of letting banks do the things that banks do. 

Everything is seating charts

So there's this:

In London, WeWork provides about half of the 8-to-10 square meters per person recommended by the British Council for Offices industry association. The company's recently opened space in the Waterloo district, the largest co-working facility in the world, has a 6,414-person capacity, according to a document WeWork emailed to brokers. That equates to less than 4.1 square meters a person, according to Bloomberg News' calculations. That's roughly the size of two standard doors laying side by side. …

WeWork offers generous common areas bedecked with everything from skate ramps to basketball hoops. Upstairs, companies renting private desks will typically work in glass-partitioned offices. Those spaces don't feel as cramped as they might, thanks to WeWork's use of desks typically smaller than those provided by its competitors. The company provides separate telephone booths in many of its locations while fixed phone lines are an optional extra, helping it stretch space even further.

Every great tech company grows out of a visionary founder's ability to see vast societal changes before anyone else, and it is fun for me to pretend that WeWork's essential insight was "man people will pay huge valuations for anything if you call it tech," but realistically the insight at the core of WeWork's success is just that it is increasingly socially acceptable for knowledge workers in stressful high-value jobs not to get offices. You know, just, look at the sheer real estate on "Mad Men," all the sprawling private offices with sofas and bar carts. In the modern economy the creative geniuses behind billion-dollar companies get five linear feet of desk space in a big open room, but, ooh, there's beer at 4 o'clock. 

Oh while I have you here on WeWork, here is a fun little tidbit from the We Co.'s initial public offering filing. WeWork had a net loss of about $900 million in the first six months of 2019, on revenue of $1.5 billion and expenses of $2.9 billion. Keen observers will note that $1.5 billion minus $2.9 billion is negative $1.4 billion, while WeWork only reported a loss of $900 million. Where did the extra $500 million come from? Well, substantially all of it—$486.2 million—came from "a gain on the change in fair value of related party financial instruments," which is explained quite deep (pages F-104 to F-106) in the notes to the financial statements.

Basically what happened is that last year WeWork sold affiliates of SoftBank Group Corp., its biggest shareholder, some warrants to buy more WeWork shares. These warrants were weird in a number of ways, but one way in which they were weird is that changes in fair value of the warrants are marked to market and flow through WeWork's income statements. Normally when a public company issues warrants they will not have any effect on income, but for technical reasons WeWork's warrants do.[7]

What that means is that, when WeWork's stock goes down, it has income under U.S. generally accepted accounting principles (the warrant, which is a liability, is worth less), and when the stock goes up it has a loss. And WeWork's stock apparently went down in the first six months of 2019, sort of. (It's hard to measure since it's not a public company, but SoftBank invested in WeWork at a $47 billion valuation early in 2019 and then bought some more shares in a tender offer at a $20 billion valuation a few months later, and surely one reasonable way of interpreting those transactions—not the only way!—is that the stock went down.[8]) So part of how WeWork made money, in 2019, was by having its stock go down. Weird!

It's just weird; it won't carry through to when WeWork is public (the warrant was exercised in July, so it's gone), and it's not like investors are evaluating WeWork primarily on its six-month GAAP financial results. I will say, though, that people made a ton of fun of WeWork last year for using a non-GAAP financial metric it called "community-adjusted Ebitda," which just sounds made up. I defended them a little; the measure was an (imperfect, amusingly named) effort to get at the underlying unit economics of WeWork's actual business. (In the IPO prospectus it is gone, but they use "contribution margin" to cover a related concept and, you know, it's fine.) But surely community-adjusted Ebitda—the cash flows from WeWork's actual buildings, plus or minus some stuff—is a better reflection of its economic reality than its GAAP net income, which includes $486 million of income to reflect the fact that its stock went down.

ICO Rating

I had never heard of ICO Rating until this week, but if you had just walked up to me and asked "what is ICO Rating" I would have said "uh probably it rates crypto tokens and initial coin offerings." And if you had asked me "does ICO Rating get paid by issuers of crypto tokens to rate their tokens more highly?" I would have said "oh gosh almost certainly yes, surely that is how everything in crypto works." Wouldn't you have assumed that? Wouldn't everyone have assumed that? Anyway:

The Securities and Exchange Commission announced [Tuesday] that Russian entity ICO Rating has agreed to pay $268,998 to settle charges that it failed to disclose payments received from issuers for publicizing their digital asset securities offerings.  

The SEC's order found that between December 2017 and July 2018, ICO Rating produced research reports and ratings of blockchain-based digital assets, including "tokens" or "coins" that were securities, and published this content on its website and on social media. ICO Rating billed itself as "a rating agency that issues independent analytical research," and stated that its mission is "to help the market achieve the necessary standards of quality, transparency and reliability." However, ICO Rating failed to disclose that it was paid by certain issuers whose ICO offerings it rated.

Yeah I mean sure of course. It is a weird situation in that (1) this is fraud, (2) no reasonable investor would have been deceived by it, but (3) the SEC should punish it anyway. The fact that no reasonable investor would have been deceived by this is bad; it is true only because the reasonable attitude toward crypto market participants is absolute cynicism. You don't want the market to work like that! If you are a participant in the crypto world and want it to grow and become mainstream, what you are hoping for is that one day, when a company calls itself "a rating agency that issues independent analytical research," there's a fighting chance that that's actually what it is. Total cynicism is the right move now, but it is not sustainable forever; if the market is going to mature it needs to have some trust.

Speaking of market making

The trading business of a bank is about buying and selling securities from and to customers. A big part of the job is just knowing who has what and who wants what; if you know that Client A is desperate to unload some bonds and Client B is desperate to buy some of the same bonds, then you can buy low from Client A and sell high to Client B and have a nice little trade for yourself. When I lay it out schematically like that it sounds easy, but all the hard grueling work is really about befriending Client A and Client B, understanding their preferences, getting them to open up to you about what they're looking to do, building a nuanced sense of how they think and what they want. It is a game of collecting information, knowing who might want that information, and putting it to use for profit. 

Anyway here's a fun story about Burford Capital:

A senior executive at Burford Capital has been accused of unlawfully trading confidential documents for a sex tape in a $91m lawsuit filed at the High Court in London. ...

In the lawsuit, Daniel Hall, co-head of Burford's global corporate intelligence, asset tracing and enforcement business, is alleged to have supplied "sensitive" documents obtained while working for a shipping client.

Mr Hall is alleged to have exchanged the documents for "video material of a sexual nature" relating to American oil billionaire Harry Sargeant III, whose assets he was investigating for another client.

Hey, look, man, if you have incriminating information about one customer that can be traded for a sex tape that will help another customer … go to town I guess? Is that wrong? I mean, it doesn't sound great, but I don't really know what the rules here are. 

Things happen

SEC Takes Action Aimed at Proxy Advisers for Shareholders. Goldman Sachs claws its way into contention for Saudi Aramco IPO. The generational feud that rocked Apollo. "Vista Equity Partners is the most aggressive private-equity deal sponsor, according to a measure by Xtract Research." Hedge Funds Have Already Bled $55.9 Billion This Year. Fannie and Freddie Plan Is Likely Released Next Month. Taylor Swift's asset restructuring. Stephanie Kelton profile. Jay Alix vs. McKinsey, continued. The Finance Minister Who Is a Star—In the Kitchen.

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] Well, some of the stuff will be reserves at the central bank, which are genuinely riskless. But that is generally a small minority of the stuff. A banking system where banks took in money and invested it only in reserves is certainly possible, but it's not the one we have now.

[2] It is a dull cliché at this point to say that making mortgage loans does seem to have caused the financial crisis, while the sorts of "proprietary trading" banned by the Volcker Rule did not, but here we are.

[3] This is a useful shorthand but also wrong. People in actual market making businesses spend tons of time and energy trying to get their customers to want to do the trades that the banks want the customers to want to do! As I once wrote: "Banks are in the business of doing trades for clients, and the way they conduct that business is by dreaming up trades for the clients to do and then pitching those trades to the clients. But they're still client trades. They still -- in the words of the Volcker Rule—reflect 'the reasonably expected near term demands of clients.' They are designed to make money because clients pay for them, not because the underlying securities will go up. That difference -- where the profit expectation comes from—is what distinguishes proprietary trading from customer-facilitation trading. But of course they are still ideas dreamed up by the banks, and pitched by the banks, and designed to make the banks money, and that can lose the banks money if the underlying securities go down. "

[4] Similar, though less controversial, was an exemption for underwriting. A thing that investment banks do a lot of is buy huge blocks of stock or bonds, for their own account, from companies and turn around and sell them to investors. This is called "underwriting" and is a basic function of investment banks. In the olden days, commercial banks were specifically prohibited from doing this, but if you tried that in 2009—when the biggest underwriting firms were all part of big universal banks—there'd be no more stocks or bonds, so it was not on the table. Instead underwriting was exempted from the proprietary trading ban.

[5] This means: The small bank goes out to the swap market and buys an offsetting swap. This, by the way, is technically how the thing gets swept up into "prop trading" in the first place: Both the new and old Volcker Rules work (slightly differently) on an assumption that trades under 60 days might be prop trades and thus illegal, while trades over 60 days are, you know, investments or whatever and thus fine. Doing a five-year swap alongside a five-year loan is thus by default not a prop trade, but selling a five-year swap and immediately offsetting it by buying the opposite five-year swap in the swap market arguably is, so the small banks needed a specific exclusion. (See page 85 of the adopting release for a bit more discussion of this point.) 

[6] "Most commenters supported allowing loan-related swaps," says the adopting release, which doesn't mention any commenters who opposed it, or really any arguments against it. 

[7] Ordinarily changes in a company's stock price don't count towards its income for fairly obvious reasons. Stock-related instruments, like warrants, are normally treated similarly, but there is an exception for some *cash-settled* instruments indexed to the stock. Because WeWork's warrants are cash-settled in some circumstances, they are marked to market. 

[8] The warrant also probably lost value just by time-value decay: It is closer to expiration now.


No comments