Unhappy UNFIThe basic sin of investment banks is that they are both counterparties and advisers. The goal, of a banker or a salesperson at a bank, is to develop deep relationships with clients so that they treat you as a trusted adviser and come to you to solve their problems. The way you solve their problems is by selling them a thing. The more the thing costs—the more fees you can sneak into it, etc.—the happier you are and the sadder they are. They trust you to advise them on what is in their best interests, and then you advise them to do something that is in your best interests. I do not want to exaggerate this. This is not "banks are irremediably evil." This is just, like, commerce. The way you get paid, as a bank, is usually that your client pays you. The more they pay you the more money you have and the less money they have. This is not that different from most other businesses: A car salesman wants to solve your transportation problems by selling you a car, and it's in his interest to add as many expensive optional features as possible to the car. But people, for both good and bad reasons, get madder about it when it's a bank. To be fair they don't usually get mad about it. Usually you buy a car and you like driving it and you are perfectly content that the salesman made some money selling it to you. Usually companies sell bonds or do acquisitions, and they are happy to have the money or the acquired company, and they are perfectly content that the bankers made some money executing the deal. But when things go wrong—when a company is not happy with its bonds, or its acquisition—everything kind of feels like a conflict of interest. United Natural Foods Inc. hired Goldman Sachs Group Inc. as its trusted adviser to do a "bet-the-company" acquisition of Supervalu Inc.[1] (Disclosure, I used to work at Goldman as an investment banker; before that, as a lawyer, I helped Supervalu do another merger.) In order to pay for that acquisition, UNFI needed to borrow a lot of money. It also hired its trusted adviser, Goldman, to be the lead lender on that loan. Presumably UNFI asked its trusted Goldman bankers questions like "what is the best way to pay for this thing" and "what mix of bonds and loans should we use" and "how should we structure the loan syndication" and all the ordinary corporate financing questions that you'd expect a company to ask of its bankers, and it was satisfied that Goldman was the right bank to lead the loan, and so it signed a commitment letter with Goldman. The commitment letter said that Goldman and UNFI had an "arms-length business relationship." It was "not intended to create a fiduciary relationship among the parties." UNFI agreed not to "claim that [Goldman] have rendered advisory services of any nature or respect with respect to the debt transactions contemplated hereby." Goldman's interests "involved interests that differ from [UNFI's] interests," and UNFI agreed that Goldman had "no obligation to disclose such interests and transactions" and promised not to "assert any claim based on 'actual or potential conflicts of interest.'" There were a lot of unpleasant words, buried in the legal boilerplate of the commitment letter! If you just read the commitment letter, you would think that Goldman dispassionately operated a money store, and that UNFI showed up at the money store looking for money, and Goldman agreed to give it the money, and they were otherwise strangers to each other, each looking out for their own advantage. On the other hand if you were an UNFI executive who talked every day to your trusted adviser at Goldman Sachs, when he sent you that commitment letter you might have given it only a cursory skim. "Oh, boilerplate, arms-length business relationship, blah blah blah," you might have said; "that doesn't mean us, we are different, our Goldman adviser loves us and we love him." The way loans like this work is that Goldman was on the hook to fund the loan, but everyone expected it to syndicate the loan to other investors before funding. UNFI agreed to help with the syndication, by providing financial statements and making its executives available to talk to investors and so forth, and Goldman agreed to try to get the syndication done in a reasonable way. (Which was in its interest, since otherwise it would have to come up with the money.) There were, as is common, "flex provisions": If Goldman couldn't get the syndication done at the interest rate it had agreed to, it could raise the interest rate, by up to 1.5%, and UNFI couldn't object. Goldman also got various fees—for structuring the loan and of course for advising on the merger—and, if the syndication had problems and it had to use the flex provisions, there would be more fees. Things became fraught—UNFI had to raise its bid for Supervalu to an uncomfortable amount to get the deal done, which seems to have made it cranky, and it seems to have blamed Goldman for this—and then the market fell and the syndication went poorly. Goldman exercised the flex provisions, adding the full 1.5% to the interest rate and demanding all the extra fees. It also apparently asked UNFI if it could have even more flex, that is, raise the interest rate even more to get the syndication done. UNFI said no, which was perfectly allowable under the contract but awkward as a matter of trusting friendship. The acquisition closed, the loan was funded, Goldman and UNFI went their separate ways, and it seemed safe to say that they wouldn't work together again for a while. The client relationship was badly soured. Then UNFI sued. Basically it didn't want to pay the extra fees, or the extra 1.5% interest, and argued that Goldman did not do a good job of the syndication and should not get all the extra money.[2] UNFI did not have a particularly good argument[3]—we talked about it last year, when UNFI sued—and last week it lost, when a New York judge dismissed its case. Here is her opinion. Mostly what is going on here is that UNFI wishes Goldman had acted in UNFI's best interests, while Goldman in fact acted in Goldman's best interests. UNFI thinks that that is bad and should not be allowed, but the judge, quite correctly, says, well, that is literally the agreement you signed. Goldman's "profit seeking motive is not improper," she writes, about Goldman's decisions to charge more money rather than less. For instance UNFI would prefer not to pay the higher flexed interest rate, which is reasonable enough, but the judge says, look, you signed a contract that said Goldman could flex the interest rate. Triggering the flex provisions dramatically increased UNFI's cost, but was financially even more lucrative for defendants. … UNFI's position concerning the implied covenant of good faith and fair dealing construes the covenant so broadly as to nullify the express terms of the Flex Provisions and create independent contractual obligations. A plain reading of the Flex Provisions indicates that the Lead Arrangers, without UNFI's consent, had the authority to exercise the Flex Provisions so long as they reasonably determined that such changes were necessary for a Successful Syndication or that such Successful Syndication had not or could not occur by the Closing Date. This is not denying plaintiff the fruit of the contract; this is precisely what plaintiff bargained for.
If Goldman couldn't sell the loan at the original interest rate, it could have just kept the interest rate the same and kept the loan on its books, or sold it to other investors at a loss. That would have been super nice of Goldman. If you had built a long and trusting relationship with Goldman, a relationship that felt like a friendship, you might have expected Goldman to do that, maybe, a little bit. Goldman did not do that. That would have cost Goldman a lot of money. Instead it raised the interest rate the maximum amount and charged extra fees. This was a totally rational thing for an economic counterparty to do, and was quite explicitly called for by the commitment letter—which after all included the flex provisions!—but it hurt UNFI's feelings. That said there are also more technical arguments about whether Goldman was allowed to charge certain extra fees.[4] If Goldman couldn't start the "marketing period" for the loans by a certain date (because, e.g., UNFI hadn't provided the necessary financial statements), then its job would be harder and it could charge extra fees. Goldman did start marketing the loan before that date, but some of the financial statements didn't come in until later, so there was a dispute about when the marketing period started and whether it earned those fees or not.[5] This dispute is mostly boring and the judge resolved it in Goldman's favor, but her discussion does include the incredible statement that "Defendants [i.e. Goldman] opine that UNFI is confusing marketing with the 'Marketing Period,' and thus, reject UNFI's argument that the Marketing Period means when marketing is done." That gives you a little flavor of where UNFI is coming from here. See, Goldman told UNFI (and the court), the "marketing period" doesn't mean the period when Goldman marketed the loan. Common mistake! The term "marketing period" actually means that Goldman charges more fees. Oh well. Anyway the upshot of all of this is that when a loan deal like this goes poorly, Goldman is allowed to look out for itself rather than sacrificing its own interests to make the client happy. But the other upshot is, hoo boy, the client is not happy. "Regardless of this decision, our complaint detailing the bank's inappropriate treatment of a trusting client speaks for itself," said UNFI after the decision. What happened here is not that Goldman cheated UNFI and UNFI sued for its money back; what happened here is that Goldman treated UNFI like an arms-length counterparty and UNFI's feelings got hurt. So it sued to express its feelings, and now it feels better, even though the lawsuit didn't go anywhere. Negative oilOops: Syed Shah usually buys and sells stocks and currencies through his Interactive Brokers account, but he couldn't resist trying his hand at some oil trading on April 20, the day prices plunged below zero for the first time ever. The day trader, working from his house in a Toronto suburb, figured he couldn't lose as he spent $2,400 snapping up crude at $3.30 a barrel, and then 50 cents. Then came what looked like the deal of a lifetime: buying 212 futures contracts on West Texas Intermediate for an astonishing penny each. What he didn't know was oil's first trip into negative pricing had broken Interactive Brokers Group Inc. Its software couldn't cope with that pesky minus sign, even though it was always technically possible -- though this was an outlandish idea before the pandemic -- for the crude market to go upside down. Crude was actually around negative $3.70 a barrel when Shah's screen had it at 1 cent. Interactive Brokers never displayed a subzero price to him as oil kept diving to end the day at minus $37.63 a barrel. ... It was only later that night that he saw on the news that oil had plunged to the never-before-seen price of negative $37.63 per barrel. What did that mean for the hundreds of contracts he'd bought? He frantically tried to contact support at the firm, but no one could help him. Then that late-night statement arrived with a loss so big it was expressed with an exponent.
Yeah I don't think you have to pay bills like that. If you get a statement demanding so much money that they can't print the amount of money on the statement, just throw it out. This is not legal advice or anything; it is more, like, dream logic. If you're day-trading on your computer at home and you spend a few thousand bucks buying cheap oil and then you get a bill for $9 million, just take a deep breath and remember that you are probably having a nightmare and soon you will wake up and everything will be fine. I mean that basically worked for Shah: Thomas Peterffy, the chairman and founder of Interactive Brokers, says the journey into negative territory exposed bugs in the company's software. "It's a $113 million mistake on our part," the 75-year-old billionaire said in an interview Wednesday. Since then, his firm revised its maximum loss estimate to $109.3 million. It's been a moving target from the start; on April 21, Interactive Brokers figured it was down $88 million from the incident. Customers will be made whole, Peterffy said. "We will rebate from our own funds to our customers who were locked in with a long position during the time the price was negative any losses they suffered below zero."
Interactive Brokers' computers thought the price couldn't go below zero, so it told its customers that the price couldn't go below zero, so, as far as those customers are concerned, it couldn't. These quotes are all from Matthew Leising's Bloomberg story about the chaos at Interactive Brokers on April 20, the day that West Texas Intermediate crude oil futures for May delivery closed at -$37.63. We talked on Friday about how that result was a bit mysterious. It's easy to understand why someone would put in a bid to buy oil at -$37.63. Getting paid $37.63 to take something that has always had a positive value is a pretty low-risk/high-reward trade. It's harder to understand why anyone would hit the bid: Why pay $37.63 to get rid of oil? The simple first-order explanation is "panic": You really don't want to take delivery of oil, you have nowhere to put it, the contract is expiring, there are no buyers at normal prices, so you are forced to sell at whatever price is available, which happens to be -$37.63. This has plenty of psychological truth to it but is still odd: Why were you forced to sell then? Those futures closed above (positive) $10 the day before, and the day after; they had a wild 20 minutes to reach that negative settlement price, but that was sort of an anomaly. There are second-order explanations available. For instance: If you are a broker, and you have a customer who is long oil contracts, and the price of oil contracts dive, the losses might eat through the customer's margin. If the customer doesn't rapidly post more margin—because they only had enough money to cover losses down to zero, or just because they're eating lunch when the market dives and you can't reach them—you might blow them out of the position by selling their contracts. You might not be as careful in selling as they would be; your priority might be to sell right now rather than at a normal price. If the only available price is -$37.63, you sell at -$37.63 and send them a bill for the rest.[6] There is another oddity to the -$37.63 price, though. Sure, it makes sense to bid -$37.63 for oil, but it also makes sense to bid -$10. When prices fell below zero, they very quickly crashed to -$37.63 instead of hovering at -$1 or -$2 or whatever. Why didn't value investors step in and say "hey I'll buy oil for -$10" to keep the price closer to normal? The Interactive Brokers story doesn't exactly answer these questions. And it's not like retail traders at Interactive Brokers were the dominant force in the oil market or anything. But it provides some very suggestive hints at how things could have gone wrong. For instance, some of the sellers at negative prices could have been Interactive Brokers clients who were blown out for insufficient margin, because they posted almost no margin at all: For the 212 oil contracts Shah bought for 1 cent each, the broker only required his account to have $30 of margin per contract. It was as if Interactive Brokers thought the potential loss of buying at one cent was one cent, rather than the almost unlimited downside that negative prices imply, he said. … To give a sense of how far off the Interactive Brokers margin model was that day, similar trades to what Shah placed would have required $6,930 per trade in margin if he placed them at Intercontinental Exchange. That's 231 times the $30 Interactive Brokers charged.
Also it helps answer the question "why did no one post orders to buy at slightly negative prices": Interactive Brokers' computer systems didn't let them post prices below $0.01. Remember Shah "tried to put an order in for a negative price, but the Interactive Brokers system rejected it." One could imagine this generalizing; Interactive Brokers may not have been alone. Maybe some proprietary trading firm that normally trades oil got out of the market on April 20 not because of concerns about valuation or volatility, but because its computer systems found negative prices confusing. Here "computer systems" might mean complex proprietary high-frequency trading algorithms, but it might also mean, like, the spreadsheet where you type in your positions, or your broker's electronic order-entry screen. If you buy oil at -$10 a barrel you may be getting a great deal, but all your spreadsheets look weird. If not everyone can trade oil at negative prices, just as a mechanical matter, then liquidity at negative prices will be limited, and so negative oil prices will be more volatile than positive ones. Once you hit -$0.01, a lot of buyers vanish, and it is a quick trip to -$37.63. The People v. Larry FinkOne occasional theme of this column is that there are (at least) two important ways of aggregating people's preferences in modern America, two imperfect ways to represent and implement the will of the people. One is that people go to the polls and vote for representatives and senators and presidents and so forth. In that world the person that Americans have chosen to run their affairs is Donald Trump. The other is that people invest their money in the market, and giant index fund providers reflect their preferences. In that world the person that Americans have chosen to run their affairs is Larry Fink, the chairman of BlackRock Inc. "But I don't want either Donald Trump or Larry Fink to run my affairs," you say, and that's perfectly reasonable; these are both crude and imperfect mechanisms of aggregating preferences, and lots of people have different preferences. "But Larry Fink doesn't actually decide how the market allocates capital," you say, "he is just the person who sits at the top of one company that makes passive investments reflecting thousands of individual—" look yes fine you are right, he doesn't actually make the calls, but Donald Trump is also sort of a symbolic figurehead for a system made up of thousands of individual decisions too. There are two massive machines for aggregating preferences and they both have a human sitting at or near the top, tweaking the machines here and there but mostly serving as metonyms for complex emergent processes. I mean Mark Zuckerberg too; I don't mean to suggest that there are only two preference-aggregating machines. But this is a financial column, and anyway the Larry Fink machine seems to be under-discussed relative to the Zuckerberg one or, Lord knows, the Trump one. Of course the two systems come into conflict sometimes. Often the conflict that I stress is that the democratic government system has a tendency to inaction, while the financial capitalist system has a tendency to do stuff, so Larry Fink sometimes finds himself filling gaps left by Donald Trump. But sometimes the conflicts are more direct: The government wants one thing and the market wants another and they fight. Here you go: Three dozen lawmakers are pushing the Trump administration to get tough on banks and asset managers that restrict financing for oil drilling and coal mining, arguing they are "discriminating against America's energy sector" and it "must be confronted." In a letter released Friday, the lawmakers told President Donald Trump that he should punish those lenders by blocking them from participating in federally guaranteed loan programs created in response to the coronavirus, including the Paycheck Protection Program. … The push, led by Republican Senators Dan Sullivan of Alaska and Kevin Cramer of North Dakota, comes as major lenders adopt policies against financing some fossil fuel projects at the urging of environmentalists and Alaska natives. Five major U.S. banks have now ruled out financing oil projects in the Arctic, and BlackRock Inc., the world's largest asset manager, has decided to forgo investing in coal. ... The lawmakers singled out BlackRock because of its coal plans and its central role in distributing credit facilities under the coronavirus stimulus law. "Its hostility toward the American energy sector is unacceptable and should be closely scrutinized," they say.
The American people (in the form of their democratically elected representatives) love coal; the market (in the form of the decisions of the largest banks and asset managers) does not. Or whatever. You don't have to believe that a letter from some congresspeople demonstrates the will of the people, or that BlackRock's socially responsible marketing demonstrates the will of the market. But these are the machines we have, and these are the results they produce. Things happenBig Money Managers Take Lead Role in Managing Coronavirus Stimulus. Tesla Sues Over Shutdown as Musk Threatens California Exit. Saudis Deepen Oil Output Cuts in Effort to Prop Up Market. Despite Recent Bets, Fed Isn't Likely to Consider Negative Interest Rates. When United Pawned Old Jets, Bond Traders Sent a Stark Warning. Coronavirus Unravels Private-Equity Playbook for Some Retailers. StanChart and HSBC face losses on loans to rice trader. Small Firms Join Rush to Return Bailouts After Rules Revisions. WeWork's woes cause mortgage-backed bonds to tumble. Lockdown Disrupts the Part of Finance That Hasn't Gone Digital. Judge Approves Windstream's Settlement With Uniti. "International Holdings Co. PJSC, which had most of its revenue in 2019 coming from fish farming in the United Arab Emirates, has reached a market value as high as $14 billion, up from about $139 million a year ago." "When Thomas and Sotomayor each finish, there's a brief but distinct loss of ambient sound, presumably marking the moment the justices mute their lines. The end of Breyer's turn, however, contains no such loss. You can even see it in the audio visualization." "Consistent with our hypotheses, we find that bullshit ability is predictive of participants' intelligence and individuals capable of producing more satisfying bullshit are judged by second-hand observers to be higher in intelligence." 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] The "bet-the-company" description is of course a cliche but I am actually quoting it from last week's New York court decision in the lawsuit between UNFI and Goldman. Other unattributed quotes in this section are generally from the opinion, with citations generally omitted. [2] To be clear Goldman doesn't actually get most of the extra money—the other lenders to whom it syndicated the loan get most of the interest and probably a lot of the extra fees. But UNFI *pays* the extra money, anyway, and it goes through Goldman, and it's mad at Goldman, not the other lenders. [3] Remember, I used to work at Goldman and am biased here. Feel free to discount my opinion that UNFI's argument was bad, though I will note that the judge in this case also found it bad. [4] There is another dispute about whether UNFI was misled into making Supervalu a co-borrower on the loans in a way that helped holders of credit default swaps. This argument was kind of silly; we talked about it last year, and let's not do that again. [5] When we last talked about it I said I was sympathetic with UNFI's argument on this point, and having read the judge's decision I still think that? I mean the judge's decision is basically that the marketing period doesn't mean the marketing period. It's not the main economic issue in the dispute, and I can see the appeal of deciding the case all one way or the other, but I might have given this one to UNFI. [6] There are more arcane explanations available too. If you were long trade-at-settlement contracts it could be to your advantage to sell oil at negative prices, since that would drive down the price you'd pay on your TAS contracts; we discussed this last month. | 
Post a Comment