Libor transition One trick that you learn pretty early on in the derivatives business is that every derivatives trade comes with a little lottery ticket that is sometimes called "restructuring value." The idea is that if you are a bank, and you have a complicated long-term derivatives trade on with a client, there is some chance that the client will, at some point during the life of the trade, come to you and say that they need to modify or cancel it for some reason. Maybe their views have changed, maybe the trade has moved against them in an embarrassing way, maybe it has moved in their favor and they want to harvest the cash, maybe they're getting acquired and closing out their old trades, maybe new rules have made the trade less attractive, maybe they have some weird tax or accounting need, I don't know, there are all sorts of reasons that they might want out of their long-term derivative contract with you. You, meanwhile, are a bank; your derivatives trades are all hedged, and you don't care unduly about staying in or getting out of any particular trade. And you have a contract, so they're stuck dealing with you. The negotiating leverage is all on one side. And so you charge them an arm and a leg to change the deal. You generally can't know that this will happen in any particular trade. And there are some types of trade where it basically never works; there are tons of liquid derivatives products with no real leverage where, if the customer wants out of the trade, you just have to get out at a reasonable bid/ask spread. But if you have a big enough portfolio of complicated illiquid derivatives trades, you will occasionally hit the jackpot on one of these restructurings, and it can make your year. It can be worth doing all of your trades at "scratch"—with zero expected value to you, based on conventional modeling assumptions—just for the chance that one or two of them will one day result in lucrative restructurings. If you're aware of this idea of restructuring value, you see it cropping up from time to time in big financial stories. One aspect of the demise of Lehman Brothers is that Lehman had a lot of trades on with bank counterparties, and it had to (or they got to) break those trades when Lehman went bankrupt, and those counterparties had all the leverage and charged Lehman an absolute fortune to get out of the trades. Or the sovereign accounting scandal at the time Greece joined the European Union was in part a restructuring of an existing swaps portfolio in a way that improved Greece's accounting and made its bank just so much money.[1] Imagine if you worked at a bank and someone came to you and said, "you have to restructure all of your derivatives now, every single one of them, it is basically the law." On the one hand: Sounds like a lot of work! On the other hand: Restructuring every derivative! You are going to get so rich! That is … happening: The major Wall Street banks -- not to mention insurers, money managers, law firms and advisory businesses -- are all mobilizing employees around the globe in anticipation of Libor's demise. The benchmark's key role in financial markets -- at last count it underpins more than $350 trillion of mortgages, loans and derivatives across various currencies -- means they're being pulled from all corners of the industry. For [Jason] Granet, the objective is clear: ensure that what many say is one of the most significant and complex transitions in the history of modern finance goes off without a hitch at Goldman Sachs. "My job completely changed," said Granet, who now reports to Beth Hammack in treasury operations after spending the bulk of his almost 20-year career at the bank overseeing money-market funds. "It's an international, complex intellectual challenge." … At a June roundtable, Goldman Sachs, JPMorgan Chase & Co., Wells Fargo & Co. and others discussed how they're establishing oversight committees, setting up joint steering groups and recruiting employees to build out their Libor transition capabilities. Consulting companies such as Oliver Wyman & Co. and Accenture Plc are advising clients on how to best manage the shift, while law firms such as Cadwalader, Wickersham & Taft have helped craft fallback language for credit contracts that reference the benchmark. Yeah, look, I take the point, there's a lot to do and it's complicated and labor-intensive. If you're the Libor Transition Guy at a big bank, you are going to be spending so much money on lawyers and computer programmers to scour and rewrite all your contracts and code to reflect new interest-rate benchmarks. On the other hand, if you are not also the biggest profit center at the bank for the next five years, I kind of think you are doing it wrong? Basically there are forty zillion loans, structured credit products, swaps, options, etc. all tied to Libor, the London interbank offered rate, which is going away due to scandal. Those loans, derivatives, etc., will all have to be re-tied to some other interest rate. SOFR, the Federal Reserve Bank of New York's Secured Overnight Financing Rate, is the preferred substitute of U.S. regulators, but SOFR is very different from Libor: It is, as its name suggests, a secured overnight rate, while Libor is an unsecured term rate. So you have to find some way of converting from Libor to SOFR (or whatever) that accurately reflects those differences. At a high level it is easy to describe how you'd do that. There are ways of getting three-month SOFR: You could just average the daily SOFRs for three months, for instance, or build a futures market that gives you a term structure for SOFR. There are ways of comparing SOFR to Libor, during the overlap period when both exist: There are basis swaps in which one side pays Libor and the other pays SOFR, which gives you a market price for the difference between Libor and SOFR; or you could always just draw a graph of Libor on top of a graph of SOFR and eyeball how far apart the lines usually are. So you could, with some greater or lesser degree of science, come to some conclusion like "3-month Libor is 0.2% higher than average SOFR," and then rewrite your contract so that every time it used to say "Party A will pay 3-month Libor and Party B will pay 1.435%" it now says "Party A will pay a three-month average of SOFR and Party B will pay 1.235%," or whatever. But every step of that is complicated and debatable, and you have to choose which benchmarks and formulas and assumptions to use, and it is basically inconceivable that the banks won't be better at all of this than their customers are. The banks have like a zillion contracts each, while a corporate treasurer might have five; the banks are in the business of thinking about this stuff and selling it to customers, while the customers are mostly price-takers. Libor transition is going to be an expensive and difficult process for the banks but I have a lot of faith that they'll find a way to make it up to themselves. The basic job of a big investment bank is to manage financial complexity: The customers have some complexity that they can't handle, and the bank takes the complexity off their hands, for a price. This is obvious in the case of, like, mergers: Mergers are complicated transactions, and the bank will tell you how to do them. It's at least as true of structured products and derivatives, though; there, the bank is selling the customer something relatively simple (highly-rated bonds with high yields, a contract that pays off if some risk comes true, etc.) and taking upon itself the complicated work of manufacturing that thing. The more complexity there is, the more money the bank can make. When there is just a giant injection of complexity into the system, that should be a good time for banks. Is Libra a security? No, I think; yes, the Securities and Exchange Commission seems to maybe think; but never mind all of that, here's this: The scene was straight out of the era of Bitcoin mania: a man on an Amtrak train to New York speaking loudly into his mobile phone, discussing a digital token he was promoting and bragging about how he planned to pump up the price. But unknown to the crypto entrepreneur, sitting a few seats away was Securities and Exchange Commission Chairman Jay Clayton, the U.S.'s top market cop. Clayton had been growing increasingly concerned that many initial coin offerings -- a twist on an IPO where investors get tokens instead of stock -- were scams. He listened for a while, his anger building, and then stood up. "I didn't tell him who I was,'' Clayton recalled. "I just said, 'you don't want to talk about this in front of me."' The 2018 encounter, which Clayton likes to call his Acela moment on ICOs, helped frame his view that virtual coin investments, once largely the province of tech geeks and anti-government crusaders, had gone mainstream and needed to be aggressively policed under the securities laws. Obviously this is not about insider trading, but it is nonetheless a good enough story that I am going to coin a Thirteenth Law of Insider Trading for it[2]: If you are going to insider trade, or commit securities fraud generally, don't talk loudly about it on the Acela from Washington to New York. Half the people on that train are securities regulators! The other half are probably doing securities fraud too, but quietly. Second, this story really captures how I've often thought the SEC thinks about initial coin offerings. There is a good sort of first-principles intellectual case that an ICO is an entirely new form of organizing human behavior that is not analogous to existing securities and so should not be subject to existing securities law, but for a long time virtually every actually existing ICO was (1) transparently just a securities offering and (2) also super fraudulent. The SEC, after encountering its thousandth fraudy ICO promotion, was just like "look if we regulate all of these things as securities offerings, and ban most of them, the world will not be worse off." I mean that is what I always assumed, but the Acela moment makes for an even better story. On an intellectual level I sympathize a bit with the venture capitalists and crypto maximalists who think that they should have an unregulated sandbox to innovate with new financial structures, but I read about all the fraudy ICOs too, and I find them exhausting, so on that level my sympathies are with the SEC. I do think that Libra is sort of obviously not a security though? Here's Bloomberg News trying, with partial success, to get Clayton to opine on that: On the Libra proposal, a number of observers have said it resembles an exchange traded fund, an investment product that requires SEC approval. One similarity: the Facebook coin would maintain a stable value by being backstopped with a basket of currencies managed by investment professionals. Though Clayton avoided commenting directly on Libra, he noted that if something looks like an ETF and operates like an ETF, the law says it should be regulated like an ETF. That holds true no matter what it's called, Clayton added. "If we have a way to reduce the cost of payments internationally, through technology, I am all for it,'' he said. "But you can't sacrifice basic principles of securities law, and other law, to allow it to happen.'' It's not an ETF! It's not a share of a fund that owns currencies; it's just a quasi-currency indexed to a basket of currencies. And it seems clear to me that people would not buy Facebook's Libra currency—to the extent that Libra ever actually, you know, happens—as an investment, to profit from price swings or whatever, but as a currency, something with a relatively stable value that they'd use to buy stuff online. Starbucks gift cards are not a security, etc., you know all of this; something about Libra being linked to a basket of currencies rather than a single one, and something about it being linked to Facebook, has led people to lose their minds about it a little bit. (On the other hand, Clayton is quite right about "and other law": If Facebook is going to set itself up as a big old bank, it might need to get some banking licenses?) Insurance Man, the economics of insurance are fascinating. On the one hand, if you're an insurance company insuring against some risk, you'd prefer, at least in the short term, for the risk never to come true for your customers, because then you'd get to keep all their premiums and not pay anything out. But if the risk never happened at all then all your customers would stop worrying about it, cancel their contracts and stop paying you premiums. This creates weird incentives. You want the risk to be big and dangerous and salient. You want everyone to worry about it all the time, so they pay you lots of money for premiums. Then ideally you'd help your clients avoid the risk, so that you can keep more of the premiums, but basically it is a volume business and you'd rather collect more premiums and pay more claims than have fewer of each. Anyway here's a story about how cybersecurity insurers encourage their customers to pay out (insured) ransoms to hackers, rather than spending the time and money to restore files from backups. This all strikes me as pretty sensible on its own terms—the backup restorations seem much more costly, in terms of downtime and lost business, than the ransoms, and presumably the hackers know this and charge ransoms that are cheaper than their targets' alternatives—but I suppose it creates bad long-term effects. It is in each individual hacking victim's interests to pay the ransom (because then they get their business back quickly); it is in hacking victims' collective interests never to pay ransoms (because then ransomware hacking will stop being lucrative and hackers will stop doing it). But the victims collectivize the problem by buying insurance, and it is in the insurers' business always to pay ransoms, because that makes the market bigger: By rewarding hackers, it encourages more ransomware attacks, which in turn frighten more businesses and government agencies into buying policies. ... As insurance companies have approved six- and seven-figure ransom payments over the past year, criminals' demands have climbed. The average ransom payment among clients of Coveware, a Connecticut firm that specializes in ransomware cases, is about $36,000, according to its quarterly report released in July, up sixfold from last October. Josh Zelonis, a principal analyst for the Massachusetts-based research company Forrester, said the increase in payments by cyber insurers has correlated with a resurgence in ransomware after it had started to fall out of favor in the criminal world about two years ago. When a new risk comes on the scene, that's good for insurers. It's an opportunity. So they have weird incentives to spot new arcane risks early and develop them into serious risks. Still this seems like an error: One cybersecurity company executive said his firm has been told by the FBI that hackers are specifically extorting American companies that they know have cyber insurance. After one small insurer highlighted the names of some of its cyber policyholders on its website, three of them were attacked by ransomware, Wosar said. Well right you want there to be a lot of hacking, but you don't want it to disproportionately occur to your clients, because then you have to pay a lot of claims. 10-Ks If you read a company's annual report on Form 10-K, and it is pretty positive and upbeat—not, like, the numbers are big, but like the actual writing is full of positive sentiments—would that be a good sign, or a bad sign, or completely uninformative for your decision to invest in the company? On the one hand, good news is better than bad news, and you probably want a certain baseline level of optimism in your management team. On the other hand, you might imagine that good managers would be particularly focused on thinking about and communicating the possible downsides, while bad managers would remain upbeat by just ignoring existential risks. And then of course there is the realist-cynical view that 10-Ks are formulaic documents largely written by lawyers and their particular level of stylistic optimism or pessimism tells you almost nothing about the actual prospects of the company. The answer is "good sign": Our results show that positive (negative) sentiment predicts higher (lower) abnormal return over days (0, +3) around the 10-K filing date, i.e., the filing period. Both sentiment measures also predict higher abnormal return over event windows of up to one month after the filing period. This finding suggests that the market underreacts to positive sentiment and overreacts to negative sentiment in the 10-K filing during the filing period. … We next examine the relation between sentiment and future firm fundamentals. We find that positive sentiment predicts higher return on assets and higher operating cash flow over the next year, while negative sentiment predicts lower ROA and OCF. The economic significance of negative and positive sentiments are comparable to each other, suggesting that positive sentiment is nearly as informative regarding future profitability as negative sentiment. That's from a blog post by Mehran Azimi and Anup Agrawal of the University of Alabama titled "Is Positive Sentiment in Corporate Annual Reports Informative?" (Here's the related paper.) The answer is yes, and I guess that sounds obvious, but if you are a believer in market efficiency it is a little surprising in two ways. One is that it suggests that you can get above-market returns just by reading a 10-K carefully. Not even "carefully," really; you can just read the 10-K, notice if it sounds happy, and if it is you can buy the stock and expect above-market returns in the short term. The other is that it suggests that corporate management teams have a level of self-knowledge, and a willingness to communicate that self-knowledge in print, that might surprise believers in strong-form efficiency. Oh by the way the paper is largely about using machine-learning methods to read 10-Ks and notice if they sound happy, and of course if you are interested in applying these lessons at a large scale as part of a professional investing strategy you probably will want to get a computer to read the 10-Ks for you. But for me the answer to the much simpler question—if you read the 10-Ks and noticed they were happy, would that be meaningful?—was also interesting and a bit surprising. Things happen "There's even an argument that the election itself falls within the Fed's purview"!? Deutsche Bank revamps treasury unit to combat negative rates. Three U.S. bond kings wield same strategy, get same result: lag their peers. Carlos Ghosn Ran a Tech Fund—Using Millions From an Executive at a Nissan Partner. 'I Would Be Shocked': $14 Billion Illinois Bond Fight a Longshot. U.S. Companies Looking for More Experience in Their Finance Chiefs. Teaser Rates Come to Crypto as Binance Starts Lending Business. Ethereum 'Almost Full' as Controversial Coin Gobbles Up Capacity. "There's no ethical rooting for any of these characters under capitalism." Steve Jobs look-alike photo boots up new Apple conspiracy theories. Workers say wearing jeans to work is worth $5,000 to them. What happened in the Hamptons this weekend? 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] On the other hand there were for a while a lot of stories about municipalities that had done interest-rate swaps with big banks, and the swaps had moved against them, and the municipalities wanted out of the swaps, but the banks quoted large dollar amounts to break the swaps. That sure *sounds* like what I'm talking about, but I suspect it mostly wasn't. Given the terrible PR of those stories, my guess is that the banks were mostly willing to break the swaps more or less at cost, without much or any new profit for themselves. It's just that "at cost," for a hedged derivative that has moved deeply against the customer, is an embarrassingly large amount of money. [2] It has a family resemblance to the frankly bizarre Ninth Law, which says that if you are going to insider trade you should do it in a company located far away from the nearest SEC office. That's based on a real academic study finding "that the SEC is more likely to investigate companies that are closer to its offices," and "that illegal insider trading increases with a company's distance from an SEC office." I have never even begun to understand the mechanism here but perhaps it is, like, insider traders just go around talking loudly about their insider trading, and the closer they are to an SEC office the more likely an SEC lawyer is to overhear them? |
Post a Comment