The repo thing The simplest weirdest thing about the spike in repo rates two weeks ago is that there was a day when big banks could borrow money at around 2%, unsecured, overnight, and lend it out overnight, secured by Treasuries, for at least 5%, maybe as high as 10%. On Sept. 17, overnight Libor, a measure of banks' unsecured borrowing costs, was about 2.15%, up from about 2.11% the day before[1]; the federal funds effective rate, another measure of unsecured bank borrowing, was 2.30%, up from 2.25% the day before. Meanwhile the Secured Overnight Financing Rate, a broad measure of the cost of repo, that is, overnight borrowing secured by Treasuries, was at 5.25%, up from 2.43% the day before. The 99th percentile SOFR borrower paid 9%, and rates as high as 10% were reported. The cost of unsecured borrowing for banks went up a little; the cost of secured borrowing against Treasuries shot up a lot. The world should not work that way: If big rational financial actors have the capacity to borrow at 2% and lend risk-free at 5% or 10%, they should do that all day long, but really they should only be able to do it for a few minutes until the rates converge. Like there's a lot of money over here, not enough money over there, you connect a pipe between them and collect a fee until the amounts level out, I don't know, this is kind of how finance works. But that's not how it worked on Sept. 17, and repo rates only got back to normal (SOFR on Friday was 1.82%, versus 1.83% for fed funds and 1.83% for Libor[2]) after the Federal Reserve intervened by setting up its own pipe and lending in the repo market. A lot of people have talked about a lot of causes for this, and I will list some here as long as you promise not to email me to say "you are missing the most important cause which is that Freedonia's central bank was liquidating its Bitcoin portfolio" or whatever. Post-crisis liquidity regulation, and real economic needs for liquidity, have made it more important for banks to hang on to their money, so they might not want to lend out money even at very attractive rates. Leverage regulation has made it more important for big banks not to borrow, so they might not want to borrow to get more money to lend out at very attractive rates. The repo market is much bigger than the fed funds market, so if you actually tried to suck money from fed funds to repo you would quickly use it all up. "Banks" in common usage are actually bank holding companies, and the actual banks that do the borrowing have limits on their ability to move money over to their affiliated broker-dealers that do the lending. Lots of other stuff too; the essential nature of the problem involves the absence of a pipe where a pipe should be, which means that its explanation will be sort of specific and finicky and plumbing-related. But here I want to abstract away from that stuff and just focus on the big simple weird thing—that during the repo spike banks could still borrow at 2% even as they were lending at 5 or 10%—and what it tells you. One very very simple thing that it tells you, one that I am almost embarrassed to mention here because it is so obvious, is that the repo spike was not an indication of a crisis for banks. The fact that the big banks could, at the height of the repo spike, borrow unsecured—without collateral, based only on their own financial strength—at around 2% means that investors just were not particularly worried that banks were running into trouble. The repo spike was bad for (some) hedge funds and non-bank broker-dealers, people who fund themselves in the repo market and don't have access to bank-type borrowing. But the banks were fine. "Fine" isn't quite the word. The second thing to notice is that the repo spike was an indication of a big opportunity for banks. Like, sure, the fact that banks could borrow at 2% and lend at 10% was not arbitraged away, which means that not enough banks were doing it, but the fact that those rates printed means that somebody was doing it. Bloomberg's Yalman Onaran and Shahien Nasiripour report that one of those somebodies was JPMorgan: Behind the scenes, lenders such as JPMorgan Chase & Co. have begun offering more money, lured by much higher returns than they get parking it at the central bank, according to executives in the market. The result: Banks have offered enough to burnish their earnings. Yet it's too little to let the market function as quietly and smoothly as it has for the much of the past decade. The dysfunction has kept the Federal Reserve pumping in money to ease the pain of hedge funds and broker-dealers that need to finance their Treasury holdings. ... When rates recently peaked at 10%, the gap with fed funds was 8 percentage points. A bank providing $40 billion could make $8.8 million in a day. That would exceed $60 million in a week. That's still dwarfed by the profits they generate quarterly in other businesses. Big banks won't start reporting quarterly results until mid-October, and in the meantime, there's no way to estimate specific profits. But market participants point to figures that they reckon illustrate the opportunity banks enjoyed. The first involves the bid-ask spread -- the difference between the price a buyer is willing to pay versus the lowest price a seller would accept. In short-term lending markets, the gap is typically slim. One hedge fund trader said he usually negotiates hard for banks to reduce their prices by just a few basis points. But as markets gyrated, he said, it became obvious banks were scoring fatter spreads, with some shaving initial offers by 100 basis points. Other participants pointed to the tri-party repo market, where banks borrow from money market funds and lend to other clients. One repo trader who works in that market described the spreads he saw as the craziest he's witnessed in three decades. Never mind unsecured borrowing: Banks could lend money to hedge funds or dealers at 5 or 10%, secured by Treasuries, and then turn around and borrow against those Treasuries from money market funds at 3 or 4%, the purest sort of pipe-connecting transaction. The money market funds mostly can't lend directly to the hedge funds or dealers,[3] but the banks can connect them and take a (generally) tiny or (occasionally) gigantic cut. The third thing to notice here is that, despite this being a big opportunity for banks, they mostly didn't take advantage of it. Some did, somewhat, but the arbitrage got nowhere close to closing, meaning that banks left a whole lot of money on the table. Partly that is for regulatory reasons: Liquidity and leverage rules limited the banks' ability to lend out the money they had or borrow more. Partly it is for genuine liquidity-risk-management reasons: Banks need to keep money around to settle customer transactions and so are hesitant to lend all of it.[4] Partly it is for sort of operational muscle-memory reasons: They forgot that this is the sort of opportunity they are supposed to seize, and how to seize it. "There are no more fed funds desks at the big banks, and the Fed staff who used to keep the wheels of the market greased are no longer there," Glenn Havlicek tells Onaran and Nasiripour. "So when there was a funding spike, nobody remembered what to do." This mostly seems like a straightforwardly bad result: Markets should clear, money should move from low-value to high-value uses, banks are in the business of making that happen, and if they can't do it then they are not doing their job and something is broken. And in particular, overnight lending secured by Treasuries is about the safest business a bank can be in, so bank-safety regulation that (1) prevents banks from doing that and (2) breaks that market is probably unhelpful. But I tend to think about the post-financial-crisis rules mainly from a cultural perspective of "have they succeeded in making banks boring?" If they have, that is not at all an unalloyed good: If banks are too boring then they will not be doing their jobs effectively, and markets might get more exciting (and scary) to make up the difference, and also of course I will have less fun stuff to write about. But banking boredom does seem to me to be the central intended consequence of post-2008 regulation. And if you want boring banks then the repo spike is almost something to celebrate: The market went haywire and created opportunities for banks to make a profit, and they mostly passed on it because (1) the rules told them not to take risks, (2) their own internal requirements told them not to take risks, and (3) they forgot how to take risks. Nice Goldman Let's say you want to borrow money and aren't sure if you are going to pay it back. Like you're borrowing it to do something risky, and if the risk doesn't pan out you won't be able to repay it; or maybe you just don't know if, when the time comes, you will feel like paying it back. Usually if you borrow money and don't pay it back, the lender will come after you for the money, but different lenders will do that with different levels of aggressiveness. If you borrow money from your parents and don't pay it back, they will occasionally drop awkward hints at family gatherings; if you borrow from a bank and don't pay it back, they might seek to garnish your wages or put liens on your property; you probably know enough not to borrow from the Mafia and not pay it back. In general if you are having doubts about your future ability or desire to repay, you should choose a lender who won't come after you very hard. Where do you think Goldman Sachs Group Inc. falls on this range? Naively you might think that the "great vampire squid wrapped around the face of humanity" would be closer to the Mafia than your parents, but nope! The delightful answer is, or at least was for a while, that in Goldman's consumer-facing Marcus business they were absolute pussycats and figured that if you didn't pay them back you had troubles of your own and they didn't want to bother you. No I'm kidding, they were absolute pussycats because they figured that if they did bother you that would look bad for Goldman, and enough things looked bad enough for Goldman that your money wasn't worth it: Executives worried that aggressive debt-collection efforts would dredge up a predatory image it had spent years trying to stamp out. So Marcus launched without a team of specialists to contact delinquent borrowers and try to recover what they owed, people familiar with the matter said. When the first borrowers fell behind, Goldman lost more money than it should have, those people said. If you were able to game this out in advance—if you were like "I'm gonna borrow $10,000 from Goldman and not pay it back, because the bad publicity of pursuing me will deter them from doing anything"—then, one, congratulations on your $10,000, but, two, send them a resume, because really you should be working at Goldman. Not in the consumer business either; they could use a bond trader with instincts like yours. The arbitrage has closed now, and Marcus "has dedicated collections staff that is specially trained, a spokesman said." I hope they're specially trained to leave a dead squid on your doorstep. (Kidding, mostly.) Disclosure, I used to work at Goldman, in the pre-Marcus days when we mostly tried to get paid back. Overstock Two of my favorite themes around here are (1) that everything is securities fraud and (2) that the goings-on at Overstock.com Inc. are pretty bananas, so it is nice to see them combined in one story: A two count federal securities fraud lawsuit against Overstock.com, Gregory Iverson and Patrick Byrne was filed Friday in Utah federal court. This follows Iverson's resignation as Overstock CFO on September 17 and Byrne's resignation as CEO on August 22. Here is the complaint, on behalf of a class of Overstock shareholders. As we have discussed a few times, Overstock declared a special stock dividend payable in shares of its blockchain-based "Digital Voting Series A-1 Preferred Stock," which would be tradable only on its own blockchain platform and not tradable at all for its first six months. Eventually the market realized that this would be a big problem for short sellers, who would be obligated to deliver shares of the blockchain stock to their securities lenders and who wouldn't be able to find any. This led to a short squeeze that drove up Overstock's stock price, as short sellers capitulated and bought in the stock. Then this squeeze deflated, as brokers agreed to accept cash in lieu of the Series A-1 stock, and as people started asking questions about whether declaring a weird blockchain dividend to cause a short squeeze might be manipulative. Overstock backed down, and the dividend was delayed and will now be freely tradable immediately. Also separately from all of this Byrne resigned as CEO to spend more time blogging about his relationship with a Russian spy. Anyway if … that … happens … is that securities fraud? Oh I don't know. The narrower questions are like (1) did Overstock announce its weird blockchain dividend for the purpose of engineering a short squeeze and (2) if it did, would that be a fraud? Specifically, would it be a fraud on people who bought Overstock stock at inflated prices during the short squeeze? From the complaint: The end of the short squeeze caused shares of the Company to immediately deflate. While shares traded to a Class Period high of $26.89 each on September 13, 2019, they traded to as low as $15.50 by September 18, 2019, three trading days later, after investors learned that the tZERO dividend was designed to be a short squeeze, and after the squeeze was first disrupted by investment banks and then abandoned by the Company. Investors also ultimately learned that it was the SEC that quietly put a stop to Overstock's attempted market manipulation scheme. Thus, while failing to disclose the true risks inherent in defendants' plan, or the true plan itself, and the real motive for the tZERO Dividend, which was to punish short sellers for a decade long campaign against them for shorting Overstock and for being a market-check, or thorn in the side of defendant Byrne. While defendant Byrne had previously, at different times, launched into public tirades over short selling and naked short selling, the tZERO Dividend was his secret plot to finally obtain hegemony over them – and it almost worked Market participants and the SEC, however, thwarted Overstock's plan but not before defendant Byrne liquidated over $102 million of his privately held Overstock shares, including over $91.98 million of stock that he sold between September 16 and 18, 2019 Oh yeah I forgot to mention that Byrne sold his stock in the middle of all this, the whole thing is so nuts. The best part is that Byrne is named as a defendant, and I have a hard time seeing him quietly settling to make this go away, so there is a real chance that the case will go to trial in the most over the top, I'm-not-out-of-order-the-whole-system-is-out-of-order sort of way. Hedge-fund education This New York Magazine/Propublica article about hedge-fund billionaire David Shaw's alleged method for getting his kids into college is really something, or honestly a lot of things. For one thing, there is a gossipy account of how Shaw's "highly compensated, Ivy-educated, hierarchical" household staff works. There is the odd irony that Shaw's wife, Beth Kobliner, is a personal-finance advice writer. The college-admissions method itself is pretty straightforward, though; you just donate a million dollars a year to a bunch of top colleges and figure your kids have to get into one of them. (It helps that the kids seem to be smart and high-achieving.) The article emphasizes the portfolio approach ("from a hedge funder's perspective, investing in multiple colleges is a classic asymmetric bet," "he's hedging his bets," etc.), and, fine, but to me the most interesting part was the pricing. Specifically the tiers of pricing: Starting in 2011, when the oldest of their three children was about two years away from applying to college, the Shaw Family Endowment Fund donated $1 million annually to Harvard, Yale, Princeton, and Stanford and at least $500,000 each to Columbia and Brown. The pattern persisted through 2017, the most recent year for which public filings are available, with a bump in giving to Columbia to $1 million a year in 2016 and 2017. The foundation, which lists Kobliner as president and Shaw as treasurer and secretary, has also contributed $200,000 annually to the Massachusetts Institute of Technology since 2013. Ouch, for Brown! (Kobliner went there, and Shaw got his Ph.D. at Stanford; neither of them went to any of the other schools.) Also what is that bump for Columbia, which came after Shaw's second child started at Yale? Are they worried that the third will need more of a backup plan? "We had to increase our annual gift to Columbia after your brother got into Yale" sounds to me like an amazingly subtle expression of billionaire parental disappointment. The best line in the story, though, goes to Jules Feiffer, which is both unsurprising (he's Jules Feiffer!) and surprising (what is he doing here?). It comes in a paragraph full of great lines: Jacob, the Shaws' youngest child, goes to Horace Mann and attended Stanford's summer jazz program for teens. Kobliner and Jacob co-authored a children's book, Jacob's Eye Patch, in 2013, the year he turned 9. Editorial guidance and illustrations were provided by Pulitzer Prize–winning cartoonist and screenwriter Jules Feiffer. ("They live in a large, makes-you-want-to-kill apartment, it's so spacious and gorgeous," said the 90-year-old Feiffer. "They offered me real money, and I was in the market for real money.") So are Harvard and Yale, to be fair. Tech-startup education Meanwhile here's a Bloomberg story about how "WeWork Was a Family Affair" in which co-founder Adam Neumann and his wife and nominal co-founder Rebekah Neumann "don't have a line at all between work and life," as she put it in 2016. "It's not even a blurred line. There is no line." She is no longer with the company, and he is no longer chief executive officer, largely for that reason. A sampling: WeWork also fostered family ties within its executive ranks. The company disclosed two connections in the IPO prospectus: One was Adam's brother-in-law, who ran the company gym. It also said an immediate family member was paid to host eight live events for the company. And there were several more instances that weren't disclosed in the filing. The chief product officer was Rebekah's brother-in-law; the longtime head of real estate was Rebekah's cousin; and for years, the company's mega summer retreats were hosted at a venue in upstate New York owned by the cousin's parents. As Adam grew older, the partying didn't stop, but he took more of an interest in his children's education. In 2017, WeWork debuted WeGrow, whose mission statement is "to unleash every human's superpowers." Rebekah became WeGrow's founder and CEO, saying she was inspired to build the school because she wasn't happy with her oldest daughter's experience in kindergarten. Students would be "raised as conscious global citizens of the world," she said. For a yearly tuition of as much as $42,000 a year, children run around the modern, blond-wood floors, staff a vegetable stand and take music lessons, in addition to more academic endeavors. The location, on the third floor of the same building as WeWork's headquarters in New York City, was designed by Bjarke Ingels Group, whose eponymous leader took a role as chief architect at WeWork last year. The school, which currently has around 100 students, was buoyed by WeWork's resources: A significant number of students are the children of employees and more than half receive financial aid, though the Neumanns' five children paid full price, according to two people familiar with the matter. Yes, another way to provide for your children's education is to just have the tech startup you control build a school for them. Doesn't Shaw's plan seem so pedestrian and unambitious and old-economy, compared to this? If WeWork's initial public offering had gone a bit better then in like 11 years it would spend a few billion dollars to build its own Yale. Meanwhile, "WeWork is formally withdrawing the prospectus for its scrapped initial public offering, capping a botched effort to go public that cost its top executive his job." And Tom Barrack's Colony Capital Inc. is "selling nearly all of its traditional holdings and channeling the money into property that focuses on the tech sector." So it's pivoting from real estate to tech? Maybe WeWork could try that, hahaha kidding sorry. Things happen Trump Officials Play Down Reports of China Investment Limits. Credit Suisse Board Backs Thiam Ahead of Spy Scandal Decision. How banks game stress tests: the 'shocking' truth. Fred Wilson: "I believe that we have seen a narrative in the late stage private markets that as software is eating the world (real estate, music, exercise, transportation), every company should be valued as a software company at 10x revenues re more. And that narrative is now falling apart." Forever 21, Teen-Focused Retailer, Files for Bankruptcy. Early French and German central bank charters and regulations. Labradoodles are fine. The hottest new psychedelic drug among trendy New Yorkers is illegal toad venom. Truro's hedgehog roundabout named best in UK. 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] We make fun of Libor a lot around here as "the interest rate at which banks don't lend to each other," but post-scandal Libor is actually very transaction-based. [2] There was a rate cut in between. [3] Why not? Partly to do with clearinghouse memberships, partly to do with credit-rating and concentration limits on money market funds, partly also, though, to do with just, like, who has whose phone number and who has set up repo agreements with whom. Banks are in the business of intermediating transactions; their job is to know which hedge funds need money and which money market funds have it, etc. The money market funds' job is not to know which hedge funds to call, so they don't. [4] Onaran and Nasiripour point out that "banks have worried about the stigma of using the Fed's overdraft facilities, where they can borrow -- or temporarily carry a negative balance -- as long as they pay it back at the end of the day. That hasn't been used since the crisis. 'No bank will ever go negative at the Fed any more,' JPMorgan Chief Executive Officer Jamie Dimon told reporters last week. 'In the old days, you would go negative during the course of the day by huge sums. So these things just change the nature of the money markets.'" If you can't go negative during the day, then you have to manage your money so as to never get below zero, which means that you can't take advantage of as many intraday opportunities. |
Post a Comment