Have you heard about the oil traders who humbled Wall Street’s top investment banks and earned $9.1 billion for their clients over just the past two years? I certainly hadn’t. The identity of these hotshots might surprise you. They don’t work at some elite hedge fund. Nor are they a handful of brilliant near-sociopaths of the type you might find in a bestseller written by Michael Lewis. Instead, they are… employees of Mexico’s finance ministry. “Uncovering The Secret History Of Wall Street’s Largest Oil Trade,” in Bloomberg Markets, is one of the most unexpected financial tales in recent memory.
The article describes how Mexico, which for years relied on oil for nearly a third of its income—making it highly vulnerable to drastic falls in petroleum prices—set out to hedge its exposure at the dawn of the new century. To say the government succeeded would be a dramatic understatement. Among other things, the story includes a near moment-by-moment account of the maneuvering in 2008 just before the financial crisis. Mexico’s team saw what the bankers couldn’t see—that prices were about to plummet—and they prepared themselves by making what looked like a crazy bet. At a moment when oil was trading at nearly $150 per barrel, they locked in a series of options contracts allowing them to sell 330 million barrels at prices ranging from $67 to $87 a barrel. That seemed bonkers—until oil crashed into the 40s before stabilizing in the 50s in 2009. Result: Mexico cashed in $5 billion at the expense of Barclays, Goldman Sachs, Morgan Stanley and Deutsche Bank. As the article makes clear, that year was hardly unusual. Perhaps the biggest mystery here is why Wall Street’s finest minds don’t have the sense that casino operators have: Don’t keep playing with the guys that take your money every time. Then again, knowing Wall Street hubris, maybe that’s not such a shocker.
Mexico has tried to keep what it calls the “Hacienda Hedge” under wraps—who’d want to expose its compliant counterparties to the sort of publicity that makes them stop taking the bets?—but Bloomberg Markets has done superb work in uncovering it (and bolstering every assertion with copious documentary evidence).
The New Redlining?
“Minority Neighborhoods Pay Higher Car Insurance Premiums Than White Areas With The Same Risk,” a co-production of ProPublica and Consumer Reports, is the antithesis of the hacienda trade story: It doesn’t qualify as surprising. But it’s well worth reading because it really delivers the goods. The reporters accomplish that by crunching all the data on auto insurance payouts for four states (Illinois, California, Texas, and Missouri—the only states that make this data available) over three-to-five year periods. Sadly, as noted, the conclusions may not surprise you, but the data is so comprehensive—anecdotes here provide the filigree on the story, not the substance—that the article is extremely persuasive. “In some cases,” it notes, “insurers such as Allstate, Geico and Liberty Mutual were charging premiums that were on average 30 percent higher in zip codes where most residents are minorities than in whiter neighborhoods with similar accident costs.” Most of the insurers were mum in response to requests to comment—though several challenged the methodology—and it leaves the impression that the old days of redlining are far from gone; they’re just somewhat better concealed.
Uber Vs. Its Drivers
There was a time when it seemed like a disproportionate quantity of technological innovation originated in the military. Everything from GPS to microwave ovens, early cell phone networks, and even duct tape are frequently credited to that source. These days it feels like the rough analogue is the videogame industry, whose ideas seem to be filtering into all sorts of areas (including the military). That thought occurred to me as I read “How Uber Uses Psychological Tricks To Push Its Drivers’ Buttons” in the New York Times. This is fundamentally a story about what the article calls Uber’s “extraordinary behind-the-scenes experiment in behavioral science to manipulate [drivers] in the service of its corporate growth.” But most of the methods are borrowed from videogames. Unlike many companies, which are constrained by laws that protect employees, Uber relies on freelance drivers and it’s free to employ all sorts of techniques to keep them on the job even when they might otherwise prefer to go home. Virtually every technique described here sounds benign on its own—for example, notifying a driver when he’s near a financial goal for the day—yet added together they end up sounding creepy and coercive.
Then there’s “forward dispatch,” a technique—similar to that of Netflix, where the next episode of a program will automatically start unless you take action to stop it—that books the next ride before the current one is completed. This feature initially had Uber drivers so mesmerized that they were forgetting to take bathroom breaks. Uber made a tweak to alleviate the problem, but, according to the article, it was only a surface fix and Uber’s manipulation efforts continue apace. These days the company is facing criticism on multiple fronts. Just in the recent past, Uber has been accused of sexual harassment and sexism, sued for allegedly stealing ride-sharing technology, described as subverting attempts to regulate it, and its CEO made an unwanted cameo in viral video in which he berates a driver. Uber has some work to do…and I don’t think it’ll find its solutions in a videogame.
Bonus: The Headline Speaks For Itself
You don’t need to read more than this headline, from the Harvard Business Review, to get the point: “Research: Executives Who Flatter Their CEOs Are More Likely to Criticize Them to the Press.” That said, there is something amusing about the use of dry scientific language to describe the timeless tendency among some to kiss the boss’s posterior before deftly slipping a shiv into his or her spine: “A one-standard-deviation increase in compliments to the CEO was associated with an average increase in resentment of between approximately one and a half and two points on a two-point scale.” It isn’t exactly Shakespeare, but I buy it. Indeed, my personal regression analysis based on years of working in offices concludes within a margin of error of plus or minus three percentage points (95% confidence interval) that a**-kissing, though not particularly noble, is in fact effective. I’ll leave it to others to conduct the peer review on that point.