Money Stuff

CDS Trades and High-Tech Managers

Also Credit Suisse, active vs. passive, breaking even and unicorn farts.


It's a little hard to believe this now, but there was a time, not too long ago, when it looked like creepy credit-default-swap stories might become a recurring feature of financial news, and might even achieve a certain mainstream prominence. "The Daily Show" did a segment on the Codere SA CDS trade, in which some hedge funds conspired with an issuer to trigger a technical default on some bonds and get paid out on their CDS contracts. "The Daily Show"! There was a brief shining moment where it seemed possible to have a national conversation about whether it is wrong for hedged bond owners to conspire with issuers to trigger technical defaults in order to get paid out on CDS contracts. 

Those were good times. I like creepy CDS stories, like that Codere trade, or the RadioShack Corp. CDS trade that went the opposite way (CDS sellers conspired with the company to avoid default). One thing I like about them is that they are a place where derivatives transactions can spill out into the real world: Zero-sum disputes between parties to derivatives contracts can cause a factory to close, or stay open. But mostly I like them for the opposite reason. They have a purity as finance; they are parables of people being rewarded just for being clever and ruthless and reading the contracts carefully. Sure, arguably the Codere and RadioShack participants made money by creating value in the real world, but it's more fun to ignore that, and to assume that the money was just a prize for being smarter than their adversaries in a pure abstract test of skill. 

Sadly, though, these stories seem to be dying out, as a genre. Single-name CDS contracts are less popular since the financial crisis, and the market has figured out some of the more obvious tricks. Late-night cable comedy has moved on from CDS structuring. But there are still a few fun stories here and there. Here's a delightful story from last week's issue of the Hedge Fund Law Report, which does not have quite the mainstream reach of "The Daily Show." Also it's about a pre-crisis trade. Still I'll take what I can get.

The story (from the Hedge Fund Law Report and this January appellate court decision) is that Bank of America Corp. put together a mortgage securitization in 2007 and sold some slices of it (tranches B6 through B12) to hedge funds run by Good Hill Partners LP. In early 2008, Good Hill also sold a credit default swap on the B6 tranche of the securitization to Deutsche Bank AG. The crisis happened and the securitization lost a lot of value. Bank of America decided to unwind it and started negotiating prices with Good Hill. Eventually they agreed to unwind Good Hill's tranches at 29 cents on the dollar. 

You might think that that would mean that Good Hill owed Deutsche Bank 71 cents on the dollar on its CDS trades. But that does not necessarily follow. The Good Hill/Bank of America deal was on six tranches of the securitization; Good Hill's CDS with Deutsche Bank was only on one tranche. Good Hill had an idea. From the court decision:

Good Hill then asked Bank of America to allocate the total purchase price for the stack of B6 through B12 notes so that the B6 notes would be paid at 100% of par, and the remaining tranches little to nothing. This action would result in no floating amount due to Deutsche Bank, as the principal forgiven would be zero. After further negotiation, Good Hill and Bank of America ultimately agreed to an 83% allocation to the B6 notes, meaning that only 17% of the principal of the B6 notes would be forgiven.

That meant that Deutsche Bank would get only 17 cents on the dollar on its CDS, instead of 71. (Bank of America didn't care: It was paying 29 cents on the dollar overall, and it didn't matter to it how the money was allocated.) Deutsche Bank complained that the purchase amounts were "allocated in a manner that appears to be potentially arbitrary and inconsistent with our understanding of the market valuation of the certificates prior to such allocation," and refused to return Good Hill's collateral. Good Hill sued, and in January it won. Good Hill and Bank of America had a deal on how much Bank of America would pay for the B6 tranche, and that was their deal, and Deutsche Bank couldn't second-guess it. Market values and economic substance were irrelevant. The CDS contract "permits Good Hill to transact business involving the B6 notes, and pursue its own interests, even if it might have an adverse effect on" Deutsche Bank. Ruthless cleverness received its reward.

There's one other amazing thing in the opinion.  Good Hill's missing collateral was $22 million, but it was awarded $90 million by the court. The extra $68 million was interest, at a rate of 21 percent per year. That seems pretty high! How did they get that number?

The relevant rate was contractually defined as equal to the cost of funds, as certified by the "relevant payee," plus 1% per annum (ISDA 2002 Master Agreement § 14). Good Hill certified that the cost of funds was 20%, the interest rate on loans from third-party investors.

The result does not change because the relevant payee was a special purpose vehicle formed by Good Hill solely for administrative purposes to segregate the funds and costs that investors incurred at the time of Deutsche Bank's breach.

So ... when it knew it was suing, Good Hill set up a special purpose vehicle to hold its litigation claims, promised investors a 20 percent return on that vehicle, and then demanded that 20 percent cost of capital back in the court case. That might be cleverer than the note restructuring, so it makes sense that it was rewarded more richly.

People are worried about weird stuff at Uber.

Are they though? This New York Times article about "How Uber Uses Psychological Tricks to Push Its Drivers’ Buttons" has an expansive idea of what constitutes a "trick":

Uber was increasingly concerned that many new drivers were leaving the platform before completing the 25 rides that would earn them a signing bonus. To stem that tide, Uber officials in some cities began experimenting with simple encouragement: You’re almost halfway there, congratulations!

While the experiment seemed warm and innocuous, it had in fact been exquisitely calibrated. The company’s data scientists had previously discovered that once drivers reached the 25-ride threshold, their rate of attrition fell sharply.

So Uber set a goal for drivers, promised them a bonus for hitting that goal, and then sent them a nice note saying "You're almost halfway there, congratulations," when they were almost halfway to the goal. Then, when they hit the goal, it paid them the bonus. What kind of trick is that? (It would be a trick if it didn't pay them the bonus.) It just seems ... I guess "warm and innocuous" is how it seems, yeah.

But it's not, is the implication, because it is "exquisitely calibrated." Setting goals, and sending nice notes to encourage people to meet the goals, is fine. But doing it based on too much calculation -- using data science to figure out what management techniques maximize effort and revenue -- seems manipulative, or tricky. "And most of this happens without giving off a whiff of coercion," says the Times, implying that the coercion is there, but high-tech filtering has rendered it odorless.

Oh boy are we going to start seeing a lot of this. For as long as there have been managers, they have tried to figure out how to motivate their workers to work faster and better. Many books have been written about the subject, and some of them have, perhaps, even been read. But the basic technique of management has always been: I'm a human, and you're a human, and I am going to try to use my understanding of how humans work to get you to work more.

But Uber -- and it is surely not alone in this -- has realized that humans don't necessarily have an advantage over computers in understanding how humans work. You can manage humans based not on your own experience as a human, but based on data. The computers know us better than we know ourselves; the algorithm controls us not with gut instinct and managerial platitudes but with rigorously optimized science. It is a little unsettling. 

Also the basic trick that Uber uses is the "ludic loop," derived from video games:

Some of the most addictive games ever made, like the 1980s and ’90s hit Tetris, rely on a feeling of progress toward a goal that is always just beyond the player’s grasp.

Tetris, sure, but that also sounds like life. The computers have been reading Camus. They've really got us figured out.

Poor Credit Suisse.

"Thiam’s Turnaround Clouded by Tax Probe of Credit Suisse" is the headline here, and I feel like there are a lot of banks, chief executive officers and regulatory problems you could substitute there. "____'s Turnaround Clouded by ____ Probe of ____"; Libor or mortgages or money laundering or take your pick. The turnaround is always happening now, the probe is always of things that happened years ago, and the sins of the past bank will perpetually be catching up with the well-intentioned present bank. (Which will presumably always be committing its own undiscovered sins to haunt the future bank, though you never know; maybe the turnaround will take.) If I were a big bank CEO, I would sometimes wish that regulators and politicians had just broken up the big banks in 2008. Wouldn't it be nice to start fresh?

Anyway Credit Suisse has a "zero tolerance" approach to tax evasion, but presumably there's some margin of error around that number.

The bank said its offices in London, Paris and Amsterdam were “contacted” on Thursday by authorities in connection with client tax matters, and that it’s cooperating. The move was so secret that not even Switzerland’s authorities knew about it. Investigators in the Netherlands arrested two people -- seizing a gold bar, paintings and jewelry -- and are probing dozens more suspected of hiding millions of euros in Swiss accounts, the Fiscal Information and Investigation Service said Friday.

I suppose if investigators brag about seizing gold bar, the probe can't be that big, but it's too early to be sure.

Active vs. passive.

Congrats everyone, or almost everyone, or almost half of everyone:

Some 45% of all U.S.-based actively managed stock, bond and other mutual funds were beating their benchmark indexes as of Feb. 28, Morningstar said.

That is good news, up from 31 percent last year; "the last year even half of all active funds beat their benchmarks was 2009." But "some investors are optimistic that conditions are right for active managers’ resurgence to continue, eventually slowing the flow of money out of actively managed funds into lower-cost index-tracking funds, a trend that has hounded many of them in recent years."

There are two basic models for how the rise of passive investing will affect active mutual funds. One is a cyclical equilibrium model in which underperformance by active funds will push people into passive funds, which will push valuations away from fundamentals, which will create opportunities for active investors, which will lead to outperformance by active funds, which will push people back into active funds. (And then vice versa.) The other is a secular death-spiral-y model in which underperformance by active funds will push people into passive funds, which will leave only the best active funds competing for the same opportunities, which will drive down performance and drive up costs for those active funds, which will lead to further disappointing performance by active funds, which will push even more people into passive funds.

Two things that the cyclical model has going for it are:

  1. It's kind of how everything else in financial markets works, and
  2. There is some limit at which it has to be true: When the second-to-last stock picker disappears, the last one is going to have a lot of new opportunities.

We'll see. But it is very plausible that the last near-decade of post-crisis economic recovery, relative stability and accommodative monetary policy has been good for passive management, while making stock-picking less valuable. The next few years of ... you know ... whatever this is ... will be a good test of whether active management's decline was cyclical or secular. Will the new environment reward bold stock-picking? Or will it just present new ways to be expensively wrong?

Breaking even.

Here's a story about how, after making capital expenditures and returning almost $37 billion to shareholders, four giant oil companies -- Exxon Mobil Corp., Royal Dutch Shell PLC, Chevron Corp. and BP PLC -- "didn’t make enough money in 2016 to cover their costs." The culprits are low oil prices, but also those $37 billion of dividends:

For companies once known as profit machines—whose executives were hauled before Congress in 2005 to explain their enormous earnings—their cash problems demonstrate just how unprepared they were for a historic crash and tepid recovery in oil prices. They have maintained the same large dividends they had when oil prices were over $100 a barrel, piling on debt and selling off assets to prioritize shareholders above all else.

Is that unusual? The average company in the Standard & Poor's 500 Index spends 101.3 percent of its operating earnings on dividends and stock buybacks. (That number is from Yardeni Research, Inc.; excluding energy companies, that number is only 95.7 percent.) So most big public companies don't make enough money "to cover their costs," if you include returning cash to shareholders as one of their costs.

People are worried about unicorns.

Elon Musk, who is currently employed as the chief executive officer of a public company that makes cars and solar panels, the CEO of a large private company that makes space rockets, and the founder of a startup that will rewire the human brain, spent some time last week drawing a farting unicorn in a Tesla. I mean, the unicorn isn't in the Tesla. He drew it in the Tesla. Tesla now has a sketchpad app. For drawing farting unicorns. Which is convenient. If running three companies and driving a car doesn't fully occupy your attention, why not use your car to doodle a unicorn? Perhaps he was parked, or in autopilot mode. I don't know. Why would you ever need to draw anything in your car? The unicorn is sort of cute.

Elsewhere here's Dan Lyons on "Jerks and the Start-Ups They Ruin":

The bro C.E.O. does what you’d expect an immature young man to do when you give him lots of money and surround him with fawning admirers — he creates a culture built on reckless spending and excessive partying, where bad behavior is not just tolerated but even encouraged.

And here is Tad Friend on "Silicon Valley's Quest to Live Forever," another good reminder of the close linkage between modern high technology and ancient mythology. In this case, vampire mythology. "If Silicon Valley billionaires end up being sustained by young blood," says Friend, "they will satisfy an ancient yearning." "This is not about Silicon Valley billionaires living forever off the blood of young people," says the founder of Google Ventures, but I guess he would say that.

Things happen.

Brexit Could Disrupt Millions of Expats’ Lives. Marquee Offerings Help Bring Life Back to IPO Market. Trump Plans Have Deal Makers Dreaming Big ($100-Billion-Cash-Takeover Big). China's Plan to Create New Shenzhen Triggers Speculative Rampage. China Has Its Worst-Ever Start to a Year For Defaults. The Rising Retirement Perils of 401(k) ‘Leakage.’ Fed's Rebel Defends Autonomy as Trump-Molded Central Bank Looms. Who’s Worth What at the White House: The Financial Disclosures. "On the same day the stockbroker for then-Georgia Congressman Tom Price bought him up to $90,000 of stock in six pharmaceutical companies last year, Price arranged to call a top U.S. health official, seeking to scuttle a controversial rule that could have hurt the firms’ profits and driven down their share prices, records obtained by ProPublica show." Donald Trump’s Ferrari Disappoints at Florida Auction. Narrower, Deeper, Older. The Brasher Doubloon. Annoying bald eagles. Industrious badger.

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Thanks! 

(Corrects description of RadioShack trade in the first item to reflect that CDS sellers, not buyers, worked with the company to avoid default.)
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Matt Levine at

    To contact the editor responsible for this story:
    James Greiff at

    Before it's here, it's on the Bloomberg Terminal.