Transition managementBefore he went to prison last week, Ross McLellan, a former State Street Corp. transition management executive, did a series of interviews about his downfall with Kip McDaniel at Institutional Investor. The way transition management works is that some client—a pension fund, etc.—wants to sell some giant portfolio and buy some other, similar yet different giant portfolio. They will go to a transition manager like State Street, which will do the buying and selling for some infinitesimal fee, one or two or sometimes zero basis points of the portfolio. What does "do the buying and selling" mean? It might mean that you take all of the client's bonds, poll banks for the highest bids for them, sell them to the highest bidders, take the money, poll banks to get the lowest offers for the bonds the client wants, and buy the new bonds from the banks with the lowest offers, and deliver the new bonds to the client at the price you got. You act as a pure agent, finding the buyers and sellers, and your compensation is your fee of one or two or zero basis points. Or it might mean that you call up your own firm's bond trading desk, sell the client's bonds to your own desk, and then buy the new bonds from your desk. You buy the bonds from your desk at a reasonable price, of course; you are not a monster, you don't let your desk rip your client off. But there is some range of reasonable prices. If a bond is bid at 99 and offered at 99.5, your client will buy it at 99.5, or perhaps 99.375 or 99.625. If your desk bought it at 99—if it bought at the bid and sold at the offer—then your firm makes 0.5 (or 0.375 or 0.625) on the trade. Plus your one or two or zero basis points. Your firm acts both as agent (you, doing the transition management) and as counterparty (your trading desk, dealing bonds); you do the trade with your balance sheet and collect a markup. How do you decide which to do? The short answer, for State Street, during McLellan's tenure, was basically that if the clients are paying attention you act as an agent and if they're not you act as a counterparty. Or, more technically, if they are covered by ERISA you act as an agent and if not you act as a counterparty: "Let's say a client wants to move $5 billion from the Russell 3000" — assets based in U.S. dollars — "into something valued in a foreign currency," McLellan says. "You've got to sell $5 billion in U.S. dollars and move it into individual currencies. A ton of pounds, a ton of euros, yen, and so on." By the late 2000s, for most pension clients based in the U.S. — and thus subject to a U.S. law regarding financial dealings, known as ERISA — State Street would act as an "agent"; they had to have the client's interests at heart. "If you're an ERISA client, we're going to route your trade through UBS or Deutsche Bank or others," McLellan says. "If you're non-ERISA" — meaning the client is not subject to U.S. law — "we're literally going to run you over and take as much money as we possibly can from you." They would do this, he explains, by routing the foreign exchange trades through State Street's own trading desk, where, like an iffy real estate agent, they would act not as an agent but as a risk-free middleman. This standard operating procedure was sanctioned at the highest reaches of the bank, McLellan claims. "We would use terms like 'We need to extract more value from order flow.'" David Puth — then head of State Street Global Markets, the research and trading division of the bank — "pushed us to internalize as much of our client's order flow as we possibly could," lauding the transition management unit's "multiplier factor of the revenue."
Originally, that was the answer only for the foreign-exchange component of the transition. The rest of it was done as an agent: You'd sell the client's old stocks, and buy its new stocks, at the lowest price available in the market, but (for a foreign client) you'd sell its dollars and buy its euros through your own desk and collect a markup. I said above that your desk charges a reasonable markup because you are not a monster, but of course you are a little bit of a monster, and the traders on your FX desk probably are monsters, and they might be tempted, for a captive trade that you bring them, to charge an unreasonable markup: "Did you do transition management trades with FX?" Paine asked, according to McLellan's recollection. "Often," replied McLellan. "Did you get good execution?" Paine continued. "No," McLellan said. "They ripped our clients' throats out."
It is all relative, though; currencies tend to trade liquidly with tight bid/ask spreads. But then State Street figured that you could do the same thing with bonds, where markups can sometimes be larger and more opaque: The standard operating procedure became non-standard in 2009, according to McLellan. That year it was determined that the model long applied to foreign exchange trading would now also be used for bonds. Who exactly decided on — and approved — that change is still in dispute, McLellan believes. The first client impacted was not Royal Mail, but a massive Middle Eastern sovereign wealth fund, the Kuwait Investment Authority. To win a piece of transition business from the KIA, State Street "bid zero" — meaning they would take no fee. "Who works for free?" McLellan asks. "If you're a $700 billion fund, you think people are working for free for you? It's mind-numbingly dumb. Nothing's for free." The Kuwaiti fund, McLellan argues, must have understood that State Street was going to make money in other ways: namely, when it came to fixed-income transitions, by acting as a riskless principal. "That's what we did for FX. For fixed-income we did the same thing."
Look he is obviously right about that.[1] If you tell a client "we will manage your transition for a fee of zero basis points," the client must know that you are making money somewhere else, and if the client is a giant sovereign wealth fund who knows how bond trading works there is a good chance that they'll know that the somewhere else is in bid/ask spreads on the bonds you're trading. If there is one simple lesson from this case, and from virtually every other piece of financial news, it is that if people are doing something for you and charging you zero for it then they're making money somewhere else, and you should probably figure out where that is. But I do not want to over-stress this lesson; it's not like the Kuwait Investment Authority learned it from this experience. The KIA already knew all of this. Charging someone zero explicit fees and then adding riskless-principal markups isn't even fraud, really; that is just the way the world works. The bad thing is that State Street didn't always bid zero, and McLellan and his U.K. subordinates Edward Pennings and Richard Boomgaardt sometimes sort of denied that it was making money elsewhere: But it was one client — the Royal Mail Pension Plan, which pays the retirements of Britain's postal workers — that ultimately led to the trio's downfall. In February 2011 the pension fund was looking to transition a large portfolio of bonds. In an email on the 21st of that month to Ian McKnight, the chief investment officer at Royal Mail, Pennings — McLellan's man in London — confirmed that "we can do this project for a management fee of 1.75 bps of the portfolio value of £1.3bln or £227,500." McKnight responded, asking for clarification. "For the avoidance of doubt can you confirm this is your full and final transition fee including all the buying and selling required by my trades?" Pennings confirmed. "The fee includes all trading required." It did not.
Yeah if you tell the client that (1) you will charge them 1.75 basis points and (2) "the fee includes all trading required," then they won't necessarily know that you're making money somewhere else. If you're a connoisseur of trading shenanigans you might find that answer ambiguous—"the fee includes all trading required" is not quite the same as "the fee is the only profit we will make from this transition"—but you can't really expect your clients to be connoisseurs of shenanigans. That answer was enough to get Pennington to plead guilty to fraud and cooperate against McLellan, and to get McLellan convicted and sentenced to 18 months in prison. "If I was on the jury, I'd convict me," he says now. Vol sellingHere's a trade: - We spin a standard roulette wheel with 38 pockets.
- If it lands on any number from 0 through 36, I will pay you a dollar.
- If it lands on 00, you will pay me $10,000.
Should I do this trade? Sure, whatever, it has positive expected value, and if we do it enough and I can afford the dollars I should eventually come out ahead. Should you do this trade? Well, no, I just said it has positive expected value for me, so it has negative expected value for you. Thirty-seven times out of 38 you'll make a dollar, the 38th time you'll lose $10,000, if we do it enough you'll go broke. But let me put it a different way. There's a 97.4% (37/38) chance that this trade will be profitable for you. (I mean, once.) Should you do a trade that has a 97.4% chance of making you a dollar? Uh … no … but what if your risk management software is broken? What if it is built entirely around the concept of "value at risk," meaning that it is focused on the maximum amount that you'd lose 95% of the time? Ninety-five percent of the time, you will make a dollar. If you are an extremely literal user of a 95% VaR risk model, this trade is risk-free. There is no risk, 95% of the time! In fact there is no risk 97.4% of the time! The other 2.6% of the time it's a catastrophe. Good lord, here are some paragraphs: Beginning in January of 2018 the scope of the volatility investment strategy was expanded to include capped/uncapped variance swaps. These swaps trade a relatively fixed return during typical to moderately high volatility conditions for a significantly more steeply tilted and nonlinear loss function during high to very high volatility conditions and carry the risk of greatly magnified losses from extreme volatility events, such as the COVID-related volatility experienced in March or that experienced in the October 1987 Black Monday event. At the same time that the VOLTS portfolio was being shifted towards the higher risk capped/uncapped contracts the size of the portfolio was being substantially increased above that of the pre-2018 strategy. A legacy risk system was being used to measure and monitor the risks of public market instruments. It assessed investment risk using a value at risk (VaR) measurement at a (95%) confidence level on an annual basis. This is a common method of sizing active risk, which works reasonably well when the returns of portfolio investments are linearly related to underlying economic and market factors. However, it does not do a good job of assessing risk in a strategy like VOLTS with nonlinear returns, especially when nonlinearity appears largely outside the confidence interval. The VaR system would thus have reported minimal risks for the capped-uncapped strategy. To fully understand the risk of an investment like VOLTS and the potential for unacceptable losses, something more than active VaR is required. The limitations of the risk system had previously been recognized, and its replacement had been identified and was in the process of being implemented. By January, 2020, Risk Management had modelled the risk involved in the capped-uncapped strategy being utilized and called for increased attention to the very low probability but nonetheless extreme tail risk of VOLTS. By the time Public Equities began to take action to reduce VOLTS exposure in early March, it was too late. Unprecedented and sustained volatility caused by the COVID-19 crisis made it impossible to unwind the positions without considerable loss.
They come from the summary of the "VOLTS Investment Strategy Review" that AIMCo (the Alberta Investment Management Corp., which invests Alberta public pensions) released last week after losing billions of dollars on its volatility strategies (which it called "VOLTS") when the market crashed in March. They are in two parts, each of which is … fine … but which fit together horrifically. The first two paragraphs describe AIMCo's volatility selling strategy. We have talked about it a couple of times; AIMCo's strategy is technically known as "capped-uncapped variance swaps" and pejoratively known as "picking up pennies in front of a steamroller," but it is in principle a form of selling insurance against market volatility. AIMCo. would collect a stable insurance premium every month, as long as volatility stayed constant or went down or went up a little or went up a medium amount or even went up a largish amount. But if volatility went up by a truly enormous amount, AIMCo. would, uh, blow up. In March, that happened, so oops. But as I have said before, in principle there's nothing wrong with this; AIMCo. bet wrong, but arguably AIMCo.—a giant public pension with stable funding and a long time horizon—is a better bearer of volatility tail risk than, say, the banks who were paying it premiums. Huge stable public pensions probably should be selling volatility insurance to fragile banks, it makes sense, it's a good transfer of risk, I have no problem with it. But then the next two paragraphs describe AIMCo.'s risk management system, which was based on a 95% VaR. Basically you figure out the worst you can do in 95% of the cases, and use that number to think about your risk. People hate VaR but it is useful for lots of things. If you own a lot of stuff, sometimes it will go up a little and sometimes it will go down a little and sometimes it will go down a lot, and VaR is a decent one-number shorthand for understanding those dynamics. If you own a normal portfolio of normal stuff, you can sort of pretend that its returns will be normally distributed, and then VaR will be a useful summary of the shape of that distribution. But if the stuff that you own is pennies in front of a steamroller, it is useless. Sometimes it will go up a little and sometimes it will go up a little and sometimes it will go up a little and virtually all of the time it will go up a little, and then quite rarely you will be run over by a steamroller. If the steamroller is in the 5% of cases that you ignore, then the trade will look risk-free, and you will load up on it. The point of VaR-based risk management is that it works well for regular portfolios of stuff whose returns look sort of normally distributed (mostly they move around a little, they rarely move around a lot); it works less well when returns have "fat tails." The point of the capped-uncapped variance swap strategy is that it is exclusively about the tails. They just do not go together. I should say, I still don't know if AIMCo.'s volatility trades were ex ante irrational, if they were mispriced or dumb or whatever. The coronavirus crisis was big and surprising; you could have quite rationally sold insurance against volatility and just been caught flat-footed in March. AIMCo. wasn't using this VaR model to price those trades; it's not like it said "well we have a 95% chance of making money so we will do as much of this as you want and charge you a minimal premium." Still it is not good to have a risk-management model that ignores the only risk you face. WirecardThe basic cultural fact is that short sellers are mean pessimists while corporate chief executive officers are nice optimists. If you are a short seller you say mean things about a company and try to make lots of innocent mom-and-pop investors poorer. If you are a CEO you say nice things about your company and try to make lots of nice mom-and-pop investors richer. CEOs are engaged in the sort of behavior that generically makes the world wealthier and better off, even if a particular CEO's actual behavior does not make the world better off. (Because the CEO's company is evil, or a fraud.) These cultural differences are boring and obvious and well known and hardly worth remarking upon, and if you are a reasonably sophisticated person in or near the financial markets you just instinctively correct for them. When a short seller says "Company X is a fraud and will go to zero" and the CEO says "no we are great we are helping millions of people and our revenue is doubling every year," you don't think "hmm this CEO seems a lot more fun to have a beer with than this grump short seller, I bet the CEO is right." You understand that they have different roles and different incentives and you try to evaluate their arguments on the merits rather than on an ad hominem basis. Unless you are the German financial regulator Bafin, in which case you work tirelessly to embarrass yourself as much as possible? At this point the general outlines of Bafin's failures around Wirecard AG are pretty well known: Short sellers and journalists had been criticizing Wirecard for years, alleging that it was involved in money laundering and fraudulently inflating revenues, but instead of investigating those allegations, Bafin instead referred the short sellers and journalists to prosecutors for a criminal inquiry, and then banned short selling of Wirecard. Eventually Wirecard admitted that it fraudulently invented most of its revenue, and now it is insolvent and being investigated for possible money laundering. Bad regulator! But now the Wall Street Journal and Reuters have more detailed stories about Bafin's failings and they are somehow even worse. Its not just that Bafin made an honest mistake in evaluating the arguments on each side; it's that Bafin basically said "short sellers are mean pessimists and Wirecard's CEO is a nice optimist so let's shut down the short sellers." From the Journal: In February 2016, two short sellers published an anonymous dossier accusing Wirecard of participating in money laundering and fraud over years. On the day it appeared, Wirecard's stock lost a quarter of its value. Rather than investigating the allegations, BaFin targeted the accusers. Less than three months later, the agency sent a 45-page report to the Munich public prosecutor, accusing 37 short sellers trading in Wirecard stock of market manipulation according to the document seen by The Wall Street Journal. Wirecard Chief Executive Markus Braun had recently purchased €18.5 million ($21.1 million) of Wirecard stock, which "suggests that Braun as CEO was convinced of the positive development of the company he helps to run," BaFin wrote. In contrast, the short sellers' "use of various domestic and foreign banks, different financial instruments and markets suggests that they were aware their actions were wrong," it added.
Braun bought stock! The sort sellers used financial instruments! He must be good and they must be bad! Come on. From Reuters: On May 12, 2016, Bafin sent its report to Munich prosecutors, outlining the case against a "network of suspects" involved in market manipulation. The report criticised Zatarra Research for "emphasising incriminating information but nothing that spoke in favour of" Wirecard. Although factually accurate, the Zatarra report's negative findings were "misleading," it concluded. In the report, Bafin pointed to the fact that Braun, the CEO at the time, had bought shares in the company, illustrating his faith in it. Braun "is convinced of the positive development of his company," Bafin officials wrote. It followed up a few days later, forwarding an email from Wirecard's lawyers to prosecutors that accused short sellers of "acting as a pack" against the company.
"The report criticised Zatarra Research for 'emphasising incriminating information but nothing that spoke in favour of' Wirecard"? That's not … you don't have to be polite in a short research report! You don't have to point out the nice things! If your argument—your "factually accurate" argument!—is that the company is a big fraud, you don't have to balance that out by saying that they have a nice website. You can just be mean, it's okay, they're a fraud! Bafin didn't see it that way. People are worried about bond market liquidityWe talk all the time about the weird science-fiction future of stock indexing, where people will only trade index funds, active management will be illegal, individual stocks will matter only as index components, Larry Fink will control the corporate world and allocate coronavirus vaccines, etc., it will all be extremely cool and I am looking forward to it. The last few months have arguably been a setback on the road to index nirvana, not so much for the reason everyone expected ("sure indexing is great when stocks go up but when they go down you want active management," oops, not so much), but because it turns out that during a pandemic people start trading individual stocks for fun and so retail stock-picking has had a surprising renaissance. Still in the long term I am confident, I welcome our robot overlords etc. What about bonds? Right now if you are a professional investor you can buy a big slug of people's home mortgages, mortgage-backed securities trade somewhat liquidly, but you can't easily buy my mortgage. The things themselves are too small, the credit analysis is too specialized, the liquidity isn't there, you just buy a bundle that abstracts away from some of the specifics. Will bonds go that way? Eventually will it be sort of weird to buy the bonds of a company, instead of just betting on a vast swath of credit using bond exchange-traded funds? Maybe? When the Federal Reserve wanted to buy corporate bonds, it started by buying corporate bond ETFs, for legal and mechanical reasons but also because its goal was to buy generic corporate bonds, and the way you buy generic corporate bonds now is through ETFs. "Actual bonds are the weird gross inputs to the smooth good useful product that is the ETF," I wrote at the time, exaggerating somewhat to try to accelerate the arrival of that sci-fi future. Anyway BlackRock Inc., which sells a lot of bond ETFs, is looking forward to it too: The bond ETF universe could almost double in less than four years as traditional fixed-income heavyweights embrace the funds, according to BlackRock Inc. Institutional investors such as pension funds and insurance companies are increasingly warming to bond exchange-traded funds in the aftermath of March's coronavirus-fueled market turmoil, the world's largest asset manager said in a report Wednesday. First-time buyers of BlackRock's iShares fixed-income ETFs collectively added roughly $10 billion of inflows during the first half of 2020, according to the firm. The shift -- at the expense of individual bonds and other fixed-income instruments -- could help swell assets in fixed-income ETFs globally to $2 trillion by 2024 from about $1.3 trillion currently, BlackRock said.
Here's the BlackRock white paper on the topic, which argues that ETF liquidity is so much better than underlying bond liquidity that institutions are increasingly using ETFs for liquidity and asset allocation and hedging. Things happen30-Year Mortgage Rate Reaches Lowest Level Ever: 2.98%. Jilted companies hold their own against shaky would-be acquirers. LSE Faces Potential Delay in EU Review of Refinitiv Deal. Wirecard Woe Spreads as Banks Struggle to Exit Loans. Burnt Banks Claim Fraud, Fabrications in Asian Oil-Trading Web. Ernst & Young Says It Isn't Responsible for Luckin Coffee's Accounting Misconduct. Bullish US options trading suggests caution 'thrown aside.' Liquidity in Sports Betting Markets. "Once merchants of fine-quality, British- and American-made goods, as well as all things trad and Ivy, Brooks Brothers has since been accused of becoming a watered-down Italian casualwear brand." Metro North Has Its Own Super Hero "Metro-Man" Handing Out Masks. 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 talked about this case back in 2014, when the U.K. Financial Conduct Authority fined State Street for this behavior, and I said much the same thing: "Note that Client E [how the FCA referred to the KIA] wanted to pay zero. Client E was not an idiot, or not exactly: It knew that it was going to pay State Street for its time and transition-management mojo. It just didn't want to pay any explicit commission, for whatever reason, and that reason probably wasn't a good one. … There's nothing more profitable than a zero-commission trade. If you've agreed to a 1.65 basis point commission, I guess you can go ahead and charge secret markups, but you'll know that you're being a jerk by doing it. But if you've agreed to do a trade for nothing? The client knows that you're sneaking something in somewhere. They're practically asking you to back up the truck." | 
↠〖*HDvideo~〗 2020 Skyggen i mit øje PelÃcula Completa En LÃnea released 2020. story: Skyggen i mit øje PelÃcula Completa En LÃnea. ↞◬∺Skyg...
Post a Comment