The Volcker Rule Gets a Little Easier
(Bloomberg Opinion) -- Programming note: Money Stuff will be off tomorrow, back on Monday.
Yesterday the Federal Reserve issued some fairly small proposed revisions to the Volcker Rule, which I guess we have to talk about. Here’s the Fed’s memo explaining the changes, and here is the much more detailed official notice. Here is an overview from Bloomberg’s Yalman Onaran, with the title “Why Wall Street Traders Aren’t Rejoicing (Yet) Over Volcker Rule.” The revisions tinker with the Volcker Rule’s requirements—removing a presumption that securities held for less than 60 days are held for “proprietary trading,” making it easier for banks to demonstrate that their market-making desks hold only enough inventory to meet “reasonaby expected near-term demand” from customers, etc.—to make compliance simpler, but don’t really change the basic structure of its prohibition on proprietary trading at banks.
The standard criticism of easing the Volcker Rule is that it will allow banks to take more risks. This may be true of the new revisions, though it is not obvious how. But it is also sort of irrelevant. Banks are supposed to take risks. That’s what banks do. Loans, after all, are risky. They take risks with customers’ deposits. It is coherent to say that banks shouldn’t make loans—this is called “narrow banking,” or the “Chicago Plan”—but it is not a thing that most people think, or that has much relevance to actually existing banking in the U.S.
Similarly, it is coherent to say that banks shouldn’t be involved in securities markets. This is sometimes called “Glass-Steagall.” But it is also—somewhat oddly, given all the vague calls that you sometimes hear to “revive Glass-Steagall”—not a thing that most people think, or that has much relevance to actually existing big banking in the U.S. The biggest U.S. banks are also investment banks, and vice versa, which is historically a little odd. (Within my lifetime—within my career as an investment banker—some of the biggest U.S. investment banks were not affiliated with depository banks; now they pretty much all are.)
So there is no general rule that banks can’t take risks, or that they can’t take risks by trading securities. There is instead the Volcker Rule, which says that they can’t do proprietary trading of securities. “Proprietary trading” means something like: trading for a bank’s own account solely for the purpose of profiting from market moves, without some other allowed purpose.
But the other allowed purposes can create risks that are much bigger than the proprietary ones. For instance the Volcker Rule allows banks to buy and sell securities for their own account as part of market making, to meet customer demand: If customers want to sell, the banks can buy; if customers want to buy, the banks can sell. Earlier this week Italian bond prices fell off a cliff, and people went around complaining that banks weren’t buying. (Not because of the Volcker Rule—this was in Europe—but because they are capital constrained.) Imagine these two scenarios:
- Every hedge fund, mutual fund and pension fund is dumping Italian bonds because they are terrified by sudden geopolitical risk. Bank X cheerfully buys all the bonds that everyone is selling, because that is good customer service.
- Bank Y does some careful analysis and decides that the German government yield curve is too steep, so it buys a modest amount of 10-year German bonds and sells a modest amount of 2-year German bonds, hoping to profit as their yields converge.
Scenario 2 is careful, modest in size, hedged, executed in fairly safe securities, and volitional; Bank Y thought it was a good idea and did it. Scenario 1 is big, directional, risky, and done only because other people didn’t want the bonds; Bank X had no view that they were good but was just trying to be helpful. Surely Scenario 1 is riskier than Scenario 2. But Scenario 2 is the core of proprietary trading: Bank Y came up with a bet and implemented it. And Scenario 1 is the core of customer facilitation: Bank X just traded to meet customer demand. Buying stuff because everyone else is selling it: kind of risky!
You could say similar things for another Volcker Rule exception, for underwriting. If a bank buys hundreds of millions of dollars of securities from their issuer all at once and then turns around and tries to resell them ... sometimes that goes very wrong.
That doesn’t mean that there’s no reason to ban prop trading by banks, while allowing market making and underwriting. Banks are not purely private entities; they rely on explicit and implicit government backstops, and governments can want them to use that backing for good social purposes and not for bad ones. It’s perfectly sensible to conclude that making pure market bets is not a particularly useful purpose for a government-backed bank, but that helping customers trade securities is. (Concluding that both are good, or both are bad: also perfectly plausible!) Similarly, you could conclude that helping companies raise money by underwriting their securities offerings is a good social purpose for banks.
You could also have views about the cultural effects of having, or not having, big proprietary trading desks within banks. Those big desks often did take big risks. And if you have big prop desks, and their traders get paid tens of millions of dollars, then all the little traders at all of the market-making desks will want to grow up to be prop traders, and will take more risk to prove their trading mettle. If the whole firm is focused on serving customers, then it will just be less appealing to people who want to take market risk, which might make it less risky.
Now I am cheating here a little bit because the Volcker Rule is not purely about the concepts of “proprietary trading” and “market making” (or “hedging,” or “underwriting,” other sorts of trading-for-their-own-account that banks are allowed to do under the Volcker Rule). It is also about … risk. If you read the actual mechanism of the Volcker Rule—both the current rule and the proposed revisions—it puts a lot of focus not on the conceptual distinction between “market making” and “prop,” but rather on making sure that any trading desk that qualifies as a market-making desk has appropriate compliance policies and risk limits designed to keep it from building up too much trading risk. In fact my Scenario 1—in which Bank X buys all the Italian bonds that its customers want to sell, just because they want to sell them—probably wouldn’t be allowed under the Volcker Rule, because the desk would quickly hit its risk limits. (It probably wouldn’t be allowed without the Volcker Rule either, because it’s not like the Volcker Rule invented risk limits.) And this is in a sense even more true of the revised rule, which gets rid of some of the complexity in the old rule’s determination of what counts as allowable market-making, and basically says that if you have and enforce risk limits on your market-making desks then everything is probably fine.
It seems to me that the basic aims of the Volcker Rule were:
- To get rid of true prop trading, that is, independent trading desks at banks whose job was to take directional market bets rather than deal with client flow; and
- To give regulators a tool to check up on the flow market-making desks and make sure that they have position limits and risk management tools and a general culture of client service and carefulness.
Goal 1 was probably the most important part of the Volcker Rule, and it has been accomplished. You can check it off the list. Those desks aren’t coming back, certainly not because of minor tweaks to the recordkeeping requirements of Volcker.
Goal 2 is also sensible, though perhaps less core, but it is harder to get right. If true prop trading is bad, then you should ban it, and the regulators did, and now it is gone; it is all very simple and binary. But if regulators are going to supervise and evaluate trading desks’ risk-management processes, then they have to continually engage in that evaluation, and think about what works and what doesn’t and how to collect data and what requirements to impose. So it is not surprising that there would need to be adjustments along the way.
Last week Goldman Sachs Group Inc. and Blackstone Group LP’s GSO Capital Partners LP unit buried the hatchet over GSO’s trade with Hovnanian Enterprises Inc., in which Hovnanian agreed to default on some debt that it owed to itself in order to trigger a payment under credit default swaps that GSO owned (and that Goldman had written). I wrote at the time that, if you are Goldman in that situation,
You tip your hat, you say “well played sir,” you hand over the money, and you learn from the experience. And if the loser does take that gentlemanly view of the matter, there is no need to, you know, do all the tedious stuff. Don’t make Hovnanian actually default on some debt to itself, don’t go ask the International Swaps and Derivatives Association for a ruling that CDS has been triggered, don’t have an auction to set the clearing price of Hovnanian bonds to find a settlement price of the CDS. Just acknowledge that that is the endgame, and pay up. The CDS trades at prices reflecting the expectation that this thing will happen. ... You can settle everything based on expectation, without waiting for it to boringly play out.
Goldman was just one counterparty, though, and there were others, and one of them—Solus Alternative Asset Management—was suing GSO and Hovnanian, and so the GSO/Goldman deal was not the end of the story. But here is the end of the story:
Blackstone Group’s GSO Capital Partners and Solus Alternative Asset Management resolved a dispute over a financing plan for Hovnanian Enterprises Inc., ending an argument that had roiled the credit derivatives market.
Solus dropped its lawsuit, and Hovnanian dropped its default: It paid the overdue interest that it had been holding back from itself, within the grace period, in time to cure the default and avoid triggering CDS. Solus—and Goldman, and anyone else who wrote CDS on Hovnanian—won’t be on the hook to pay out, and GSO—and anyone else who bought CDS on Hovnanian—won’t get any payout. (Hovnanian will still get the favorable financing from GSO that it had been promised in exchange for the default.) One presumes that some money changed hands (“a person familiar with the matter said GSO was compensated for allowing Hovnanian to make the payment,” reports the Wall Street Journal), but that it was rather less than the full payout—something like 60 cents on the dollar—that would be implied by the trading prices of the new bonds that Hovnanian issued to engineer the whole thing. I tend to think that Blackstone should be rewarded for its cleverness, but when everyone from Goldman Sachs through the Commodity Futures Trading Commission to the Pope is against a trade, you can see why Blackstone might not want to cling to it to the bitter end.
Also it is a nicer resolution. Actually doing all the stuff—declaring Hovnanian in default on its debt, roping in the International Swaps and Derivatives Association to get a ruling on whether CDS was triggered, holding an auction for Hovnanian bonds to set the CDS settlement price, both sides trying to manipulate that auction, etc. etc. etc.—would be a long and contentious process with lots of potential for surprises on both sides, as well as just a lot of boring pointless paperwork just to end up with Solus writing GSO a check. This way, GSO did a clever thing, and its counterparties—grudgingly, sure—acknowledged its cleverness in the appropriate way (with money), and everyone can move on without causing too many repercussions in the real world. Except for Hovnanian. Hovnanian got some cheap financing out of it! That’s nice.
The story isn’t over in at least one respect, though, which is that everyone at least vaguely agrees that this is Not Quite What We Had In Mind, and that CDS documentation needs to be revised to make this trade impossible in the future. I do not think that will be easy, exactly, but expect ISDA to form some committees and propose some revisions. (Solus says it “looks forward to working alongside other market participants, regulators, and ISDA to improve existing CDS documentation and enhance the integrity of the credit derivatives market.”) That, interestingly, I do not expect to be an especially contentious process. Nobody is, like, on a side of this trade in the abstract; there is no natural constituency for “manufactured defaults are good.” Once you have an actual position, and an actual trade presents itself, you are going to argue for whatever makes you the most money, but behind the veil of ignorance everyone wants the same thing, which is for the CDS rules to make sense and correspond to economic reality and be robust and trustworthy and appealing to as broad a market as possible.
Oh fine and also, if you’re smart, maybe you want the rules to leave a bit of room for smart people to exploit them. But if you’re so smart you can smuggle that in without anyone else figuring out what you’re up to.
Buffett + Uber!
There is an unstoppable-force-meets-immovable-object quality to “Buffett Proposed $3 Billion Uber Investment But Deal Crumbled.” On the one hand, Warren Buffett’s schtick is pretty much “hi I will give you a lot of money but in exchange you have to give me laughably favorable terms because I am Warren Buffett.” On the other hand, Uber Technologies Inc.’s schtick is pretty much “hi we will let you give us a lot of money but you have to give us laughably favorable terms because we are the epitome of a hot tech unicorn.” (It raised money at a $62.5 billion valuation without financial statements!) The bid-ask is wide. There is perhaps a sliver of room for compromise, because Buffett tends to want laughably favorable financial terms—“Under the proposed agreement, Berkshire Hathaway would have provided a convertible loan to Uber that would have protected Buffett’s investment should Uber hit financial straits, while providing significant upside if Uber continued to grow in value”—while tech unicorns tend to insist on laughably founder-friendly governance terms, but in fact it didn’t work out:
During negotiations Uber Chief Executive Officer Dara Khosrowshahi proposed decreasing the size of the deal to $2 billion, one person said, hoping to get Buffett’s backing while giving him a potentially smaller share of the company. The deal fell apart after the two sides couldn’t agree on terms, one of the people said.
Uber, the thing is, doesn’t really need money. It has lots of money and has no evident trouble raising more. The trade here was that Uber had taken a bunch of reputational hits in quick succession, and bringing on Buffett and his folksy image might have been a good way to help solve that problem. But that’s a weird problem to solve with money. They should have offered him a $1 investment and a board seat. They wanted Warren Buffett floating around, but they didn’t want to give him billions of dollars’ worth of stock at favorable terms. And he wouldn’t float around for less. You can’t really fault either side here; both Buffett and Uber have such good track records of getting exactly what they want that it would be weird for either to settle for anything else.
“This is a sexy product that people get excited about owning and tell their friends about,” says a guy, and imagine if he was talking about a sexy product? People are always going around telling their neighbors about this hot new virtual-reality erotica experience! But yeah no, he is talking about tax minimization, which in certain circles carries its own erotic charge. In particular he is talking about private placement life insurance, a way to put money into hedge funds without ever paying taxes. The disadvantages are (1) you need at least $2 million to open an account and (2) you can’t spend the money until you die (though you can borrow against it, which is almost as good), but those are relatively minor disadvantages if you have more money than you can spend. If you’re rich and well-advised enough, you can build yourself a consumption tax system out of the U.S. income tax; if you’re paying taxes on income that you don’t spend then you’re doing something wrong.
My inbox is mostly a steady stream of PR emails about initial coin offerings, most of which I delete without reading, but I got one the other day that actually surprised me. It began:
So, you want to contribute to your first ICO? There are several barriers that exist today, including lack of education and an extremely competitive field, that prevent or deter people from contributing to an ICO.
Let’s not discuss the actual claims there, which are, you know, whatever. But the word “contribute” is interesting! Classically the reason to take your money and use it to buy tokens in an ICO is that you think those tokens will be worth a lot more than you paid for them. Classically the way someone promoting that process would describe it is as “investing.” (You could say “speculating,” or “gambling,” if you were more skeptical.) “Contributing” is … a different word. If you invest in a fund, you expect to get back more money than you put in. If you contribute to a fund, you expect to get back no money at all. “Contribute” is not a good marketing word.
On the other hand it is—and this is of course not legal advice—but it is plausibly a good legal word. If you invest in a tradable token that you expect to gain value due to the efforts of others, then that is, customarily, a security, and there’s a good chance that the people offering it to you are breaking the law by not registering it with the Securities and Exchange Commission. If you contribute to a crowdfunding effort with no expectation of profit then … look, I don’t know what to tell you, you obviously didn’t do that, there is no conceivable universe where people are contributing to heavily promoted celebrity-endorsed ICOs out of disinterested generosity, but whatever, fine, if you “contributed” to an ICO, then maybe that is the right magic incantation to avoid securities-law liability. (“They found a lawyer I guess,” I joked on Twitter, though someone replied “That, or they want it to seem like a retirement plan.”) It will be weird if this becomes the standard ICO regulatory approach: Will people still buy into hyped ICOs if the hype studiously avoids any discussion of potential profits?
There's a limited supply of corn in the world. Corn has value through “proof-of-work”: a good corn harvest is proof of a solid year's hard work. Supply and demand, something something blockchain.
Also, “corn” kind of sounds like “coin.”
Deutsche Bank’s U.S. Operations Deemed ‘Troubled’ by Fed. Jamie Dimon Tops List of Highest-Paid Finance CEOs. Tepper's $2.3 Billion NFL Purchase Comes With a ‘Wacky’ Tax Perk. SoftBank Vision Fund to Invest $2.25 Billion in GM Cruise. Ex-Goldman Sachs Banker Sues Federal Reserve Over Lifetime Ban. CryptoKitties on Morning Edition. The Internet Is Loving This Buff Cat.
©2018 Bloomberg L.P.