Money Stuff

Sugar Trades and Nudist Camps

Also search data, Ponzi schemes, email deletions and Uber's culture.

First you get the sugar.

Wilmar International Ltd., a major palm oil producer but "a relative newcomer in the sugar business," is stocking up on sugar the old-fashioned way: by taking delivery on futures contracts.

Wilmar has been scooping up sugar by physically settling tens of thousands of futures contracts and collecting the commodity from ports across South America and elsewhere. The company has bought more than 6 million tons of sugar in this manner since 2015, enough to fill roughly 3,000 Olympic-size swimming pools at a cost of some $2.3 billion.

I never quite understand size comparisons like this. When in most people's lives do they come across a swimming pool full of sugar, never mind 3,000 of them? What would you do with a pool full of sugar? I guess if you got 3,000 more Olympic-size pools full of water and mixed them with the sugar pools, you'd have 6,000 Olympic-size pools full of simple syrup. That'd be something. It would take Katie Ledecky a little over two days to swim them at her 800-meter pace, though I suppose that pace would be hard to keep up for two days, or in syrup. When she emerged, she'd probably want a drink. If you combined the 6,000 pools of simple syrup with 72,000 pools of rye and 1,200 pools of bitters, you'd end up with 79,000 Olympic-size pools of old-fashioneds, or about 3 trillion cocktails, or about 400 old-fashioneds for every man, woman and child in the world. 

Also, just, 5 million tons of sugar is 5 million tons of sugar. The average American consumes 94 grams of added sugar per day, or about 76 pounds per year, so 5 million tons would supply America for about 5 months. Putting that back in pool terms, Americans eat about 20 Olympic-size swimming pools of sugar per day, or about one lane every seven minutes.

Anyway! There's a reason that taking physical delivery of commodity futures is mainly a cautionary tale for new traders. If you want sugar for your cookie factory, the thing to do is to get a contract with a sugar producer to deliver exactly the sort of sugar you want, in the quantities and timing you want, to the door of your factory. Sugar futures are essentially financial contracts, intended to give hedgers exposure to the price of sugar, not to give people sugar in the most convenient form for baking. The sugar underlying the futures contracts exists in a sort of hypothetical financial-sugar space, which is not where you want to bake your cookies. But of course for the financial sugar to be a useful hedge to real sugar users' costs, there needs to be some porousness between the financial and cookie sugar spaces. And if you are a large global sugar user, maybe you're best positioned to create that porousness -- to collect sugar "from ports across South America and elsewhere" and ship it off to cookie factories, turning the financial sugar into real sugar. That works better in sugar than in most commodities:

Mr. Bohbot said the company has found it economical to purchase sugar in bulk using futures contracts, because the exchange’s rules require sellers to deliver the sugar on board buyers’ ships, which facilitates international trading. In other commodity markets, such as grains or metals, the handover usually happens inside warehouses in locations that often might not be easily accessible.

On the other hand, Wilmar isn't a cookie company; it is more a producer of sugar than a consumer. It seems to be buying futures sugar to augment its own production and ship to customers; its on-exchange buying "is a symptom of the growth of their business." Or maybe its financial-sugar transactions are meant to have effects mostly in the financial-sugar space:

At one point in 2015, when sugar prices were at multiyear lows because of a world-wide glut, Wilmar bought so much that traders say the company in effect mopped up that year’s global oversupply. In the rally that followed, sugar prices more than doubled.

Then, as prices peaked in September last year, Wilmar changed course and delivered excess sugar it owned to other traders on the exchange. Sugar prices fell 24% in the ensuing months.


The most charming of the many charming things about Bridgewater Associates, the world's largest and most self-reflective hedge fund, is that whenever journalists write a story about how weird it is, Bridgewater responds with a bitter accusatory denial that always makes it sound much weirder. So the other day the New York Times reported that Bridgewater had had the future director of the Federal Bureau of Investigation conduct multiple interrogations of its co-chief executive officer to find out whether she or her assistant had written an email, videotaped those interrogations, and required employees to watch them. ("Due to a leak to the media," said Bridgewater, "a mountain has been made out of a molehill.") And then yesterday Bridgewater's founder Ray Dalio spoke at a New York Times conference, criticized the story, and explained that Bridgewater's culture really isn't that weird when you think about it:

Speaking at a conference sponsored by The New York Times in Half Moon Bay, Calif., the 67-year-old billionaire likened that culture — which he calls “radical transparency” — to going to a nudist camp for the first time.

“You first walk into a nudist camp and it’s very awkward,” he said at the New Work Summit conference. For people to get beyond that, he said, “there has to be getting over the emotional reaction.”

Yes wait what? Bridgewater is not that unusual, I suppose, if your context is that it's like a nudist camp; but that's an unusual context to have. Most people are not going to work at hedge funds because they want to expose their most private parts to strangers' critical judgment. But that is exactly what Bridgewater wants. The difficulty is that Bridgewater is run by people who are very smart and rigorously logical, so everything that they do and say makes perfect sense in their context. It's just that they start from a slightly different place from the rest of us.

Elsewhere, here is a story about two former Bridgewater employees who left and "founded an online recruiting company, Scouted, which uses video interviews, written questions, and a machine-learning algorithm to match college seniors and recent graduates with employers seeking young talent." Their thing, like Bridgewater's, is deep self-reflection and critical self-understanding:

The conceit of Scouted is to then “nudge” companies into understanding the underlying personal traits a position requires, and also to nudge applicants into understanding what they fundamentally desire in a job and what they can offer an employer. The consequence is that both employers and employees obsess less about superficial factors like the brand name of a school or company and more about the intrinsic fit between a job opportunity and candidate.

“A client will be hiring for a sales role, will say, `I need someone who is really hardworking, and really smart, and a great-problem solver.’ And they give you a laundry list of everything that is important. But if you ask them to force-rank those items, nine times out of ten it is different from where they started. A lot of times the raw horsepower, the intellectual ability, is far second to having high EQ and strong diligent follow-up,” offered Ms Levine.

Sure, you say you want smart salespeople, but everyone says that. Deep down, do you really care if your salespeople are smart?

An oddity about the Bridgewater culture, which you see reflected in Scouted, is the combination of psychologically searing personal reflection with artificial intelligence. You might expect an organization with very visible individual performers -- a basketball team, a gut-driven trading firm -- to put a lot of effort into understanding its employees' psyches. But Bridgewater just produces a robot that makes investing decisions. By externalizing so much decision-making to the robot, perhaps Bridgewater employees become more conscious of the structure of decision-making, and more rational and reflective in their own decision-making. Or, alternatively, by offloading all the rational decisions to the robot, they're left with more time to get in touch with their emotions. In any case a machine-learning algorithm to evaluate someone's emotional intelligence feels somehow weirder than one that evaluates their rational intelligence. What makes the algorithm so good at judging emotions? 


Here's a story about how quantitative fund firm Acadian Asset Management has "struck a deal with Microsoft Corp. to use its Bing Predicts big data technology to inform its investment decisions." So now Acadian will know exactly what the six people who use Bing are spending their money on. No, no, I kid; "the technology now has data from millions of daily searches on Bing, the default engine across Yahoo! Inc.’s string of websites and on Apple Inc.’s Siri." I'm sure it's very useful.

Should you be mad that a big money manager is using your search data to get an edge in the financial markets? (I mean, if you use Bing for searches.) Ehh. The basic social contract of the modern internet is that it gives us amazing free stuff in exchange for collecting and using all of our information, and if that is the deal then feeding the information to hedge funds to make investing decisions is really no more offensive than feeding it to retailers to make advertising decisions. Honestly I think I might like it more? Imagine a social network that was optimized to provide actionable data to hedge funds, rather than one that was optimized for advertising. Doesn't it seem like that would be a more useful network? A calmer one? Advertisers want engagement, reaction, emotion; investors want accuracy.

Also, we sometimes talk around here about the socialist calculation problem and modern investing technology. Markets aggregate individual decisions about what to buy into signals that tell society what to produce. That is a decentralized, anonymous, circuitous process: I buy a widget, and you buy a widget, the widget company makes more money, some new widget startup is attracted to the sector, it raises money from investors who made a lot of money on their last widget investment, it builds a new factory, it rolls out new widgets, and so forth. The mid-20th-century dream of the planned economy was that there could be some technological way to just figure out what consumers want and direct producers to supply it, without the mechanism of prices and markets. If Microsoft can just figure out what people want directly (from what they tell Siri they want), and then tell Acadian (which "also manages a portion of Microsoft’s balance-sheet assets") where to invest, then that short-circuits the process at least a little bit. You can go from consumer desire to capital allocation without passing through consumer purchases

Puerto Rico.

Puerto Rico's teachers' pension fund is in bad shape:

In Puerto Rico, for instance, the pension funds are so short of cash that money contributed by working teachers basically flows straight out to retirees. None of Puerto Rico’s current teachers can expect to get their money back, because the fund is due to run out of money in 2018, long before they retire.

That is, essentially, a Ponzi scheme. But this structure is legal in Puerto Rico because of a complicated series of changes in the law brought about in recent years by the island’s financial crisis.

People sometimes make similar claims about social security, which also runs on a more or less pay-as-you-go basis, but which is not nearly so worrying. The difference is not so much in mechanics as in demographics: If you have a healthy growing economy, then each generation's retirees are supported by a larger number of working people who are producing more stuff. So it's no problem for the workers to support the retirees. If the economy is shrinking and young people are leaving, then there will be fewer people producing less stuff to support a growing population of retirees.

It is often useful to talk about these issues in terms other than money. The economy produces stuff, and workers consume some of it, and retirees consume some of it. Pension structuring and accounting, or just the amount of money in the pension fund, can affect the division of claims on the stuff, but not really the amount. The stuff has to come from somewhere. If you're making less stuff, and giving more of it to people who have retired from making stuff, then you are going to have a problem. 

Don't do this.

ZTE Corp. will pay $892 million in fines for violating U.S. sanctions in a "six-year-long conspiracy to acquire U.S. technology, send it to Iran and mask its involvement through a network of front companies." Don't do that, obviously, but if you do, probably don't write and preserve a memo weighing the pros and cons:

ZTE officials knew they were taking a major risk, according to a 2011 internal document posted earlier by the Commerce Department. One company official warned that the Iran work could “potentially put us at risk of being put on the Blacklist by the U.S.,” according to the document. If that were to happen, the company could face “the risk of losing the supply chain of U.S. products,” the document continued.

ZTE also hid data from a forensic accounting firm its lawyers had hired to review the transactions, prosecutors said, including by forming a 13-person team to “sanitize the databases” of information relating to its Iran business and using an auto-delete function to erase certain employees’ emails on a daily basis.

Lots of companies have periodic email auto-delete functions, though a one day auto-delete is a little aggressive. But the trick is, you don't want to put the auto-delete in after you start doing bad stuff, or particularly after you start being investigated for doing the bad stuff. If you are starting a business today, consider running all communications on Snapchat. That way, if you later get up to no good, no one can accuse you of switching to Snapchat to hide your misbehavior. You were doing it all along! Obviously this is not legal advice. Or business advice really; if your business is mostly legit, you might find it useful to save some communications.

People are worried about unicorns.

People are so worried about all the weird stuff happening at Uber that its chief executive officer, Travis Kalanick, is looking for a ... person ... to make it ... less weird? I don't know. "We’re actively looking for a Chief Operating Officer: a peer who can partner with me to write the next chapter in our journey," said Kalanick's statement. One presumes that this has something to do with his previous confession that "I need leadership  help, and I intend to get it." "While a wide range of candidates are being considered, sources said Uber’s board would prefer a woman exec for the job," reports Kara Swisher. "This strikes me as a glass cliff sort of situation for any woman who takes the job," tweeted Elizabeth Spiers: Reining in the CEO is not exactly a high-upside job. If you get hired as Kalanick's superego, you're not going to get much credit for any business successes, while any further weird debacles will be blamed on you, since you were hired to stop them. 

Also, what?

“Generally speaking, talent is very tight, and the pool of available executives within the technology industry is small,” said Richard Marshall, a partner with the executive search firm Korn Ferry. “And for fast-growing, founder-led companies, it’s hard to find someone with the skills needed who is a cultural fit.”

Don't they want someone who is not a cultural fit? Isn't the whole problem here that Uber's culture is bad, and damaging its business? "Culture" is such a weirdly freighted word in the startup world; every startup at some level believes that its culture is a critical part of its success, and that the most important choice in hiring is "cultural fit." That's probably even true of Uber! Being a hyperaggressive Randian firm that ignores local laws has surely helped Uber grow quickly and raise money. Swisher notes "how loyal and tolerant [Kalanick] has reportedly been to a range of what Huffington described as 'brilliant jerks' at Uber": A certain tolerance for jerks, Uber might reasonably believe, has been good for business. Now it is perhaps becoming less good, but abandoning the culture that got them here to become a normal company seems like it would be hard. Particularly if the person tasked with doing that is also hired based on cultural fit.

Elsewhere, here is my Bloomberg Gadfly colleague Shira Ovide on down-round acquisitions, though of public tech companies, not private unicorns: "In nearly every acquisition of a public U.S. tech company since 2015, the tech firm sold for less than its record high share price." 

Things happen.

Women Managers Are Winners in Market Shift to Passive Investing. U.S. Zeal in Suing Banks for Lending Bias Is Expected to Cool. Questions About Loyalty to Trump Stall Mnuchin's Treasury Picks. Lebenthal Reaches Deal to Sell Most of Business, but Famed Brand Will Be Kept. Central Banks Ratchet Up Foreign-Currency Reserves. Hammond Presents Upbeat Picture as Britain Braces for Brexit. Blue Origin, Jeff Bezos’s Moon Shot, Gets First Paying Customer. The Era of Overconfident CEOs Is Waning. Eight Fossil Fuel Majors Seen Polluting as Much as the U.S. These Are the 25 Best-Paying Jobs in America. Why the day robots pay taxes is the day we become the tools. LaGuardia is great. Creating the Girl Who Stared Down the Wall Street Bull. Will Women Revive the General Strike in America? Do dogs like cats?

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

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.