Money Stuff

Was Chicken Libor Manipulated Too?

Also atomic clocks, bond marking, an informal blockchain, crypto and flamethrowers.

Chicken Libor!

We have talked a few times before about what I like to call "Chicken Libor," the alleged manipulation of the Georgia Dock index of chicken prices. The Georgia Dock index was a sort of country version of Libor, in which an employee of the Georgia Department of Agriculture -- his name is Arty Schronce -- would call up chicken producers each week and ask them what the price of chickens was. As with Libor, there was no market check on this data, and so the chicken producers could make up numbers if they wanted. As with Libor, the chicken producers could profit by manipulating the number: They had contracts to sell chickens to big customers at floating prices tied to the Georgia Dock price, so submitting a higher hypothetical price to Arty Schronce could bring them higher real profits.

This week US Foods Inc. and Sysco Corp., two giant food-distribution companies, sued a bunch of chicken producers for manipulating the Georgia Dock price, and their complaints (SyscoUS Foods) suggest even more uncanny parallels between Georgia Dock and Libor. For instance, Libor was supposed to be the rate at which banks could borrow from each other, and banks were supposed to submit their borrowing rate in various currencies and tenors. But they often didn't borrow in those currencies or tenors -- Libor was "the rate at which banks don't lend to each other" -- and so they had no choice but to make up numbers in response to the Libor survey. The fact that they often made up numbers to help their derivatives positions isn't great, but even if they had made up numbers with the purest of motives, they would still have made up numbers.

Likewise Georgia Dock was the price at which chicken producers didn't sell 3-pound chickens in Georgia:

The price quoted by each Georgia Dock Defendant to the GDA was based on 2.5- to-3 pound whole chickens, but only a handful of the Georgia Dock Defendants actually processed 2.5-to-3 pound birds in Georgia, so, according to internal GDA documents, the Georgia Dock Defendants were “supposed to adjust their whole bird quote as if they are producing that sized bird.”

Libor's calculation mechanism also discarded the highest and lowest quotes to get a trimmed average, so that a single outlier bank couldn't affect Libor too much, and you'd need collusion between multiple banks to really move Libor. Georgia Dock did too:

Schronce, who began compiling the Georgia Dock Defendants’ price quotes in 2012 following the retirement of his predecessor, first made a preliminary calculation based on the single price quotation from each company. Then “[a]ny company that provides a whole bird quote that is more than one cent above or below the initial dock price calculation will not be included in the calculation for the whole bird dock price that week. Its voice is taken out of the formula and the dock price is recalculated without it.” This so-called “one-cent rule” is, according to internal GDA documents, meant “to shield [] one company having the ability to greatly influence the price up or down.”

The allegation, in Chicken Libor as in real Libor, is that the producers colluded to push up the index. Unlike in real Libor, there don't seem to be any electronic chats between chicken traders agreeing to manipulate Georgia Dock. Perhaps US Foods and Sysco will find some in discovery, but I have a feeling that chicken markets don't operate that way.

I love Chicken Libor's parallel to the real Libor scandal -- they each even have a Tom Hayes! -- but to be fair the lawsuits are mostly not about Chicken Libor. They're mostly about other forms of alleged collusion. The gist is that a bunch of big public chicken producers were losing money when feed prices went up, and wanted to cut production and raise prices, but couldn't do that on their own because chicken is a commodity product and no one supplier has pricing power. So their cunning plan was allegedly to cut production and raise prices, but also talk about it on their earnings calls, so that everyone else knew they were planning to cut production and raise prices and would do the same. From the complaints:

In early 2008, through press releases, earnings and investor calls, at investment bank conferences, at events hosted by Agri Stats, and at the myriad trade association meetings attended by many of their senior-most executives, Defendants began calling on one another to heed the mistakes of the past, preaching that oversupply was decimating industry profits, and increased supply-side “discipline” was needed to halt the downward trajectory of chicken prices.

This production “discipline”—i.e., the reduction or relative stabilization of industry capacity, particularly at the breeder flock level, where such efforts would be most effective—was a mechanism to increase Defendants’ profits. Defendants’ efforts were supported by public statements made by their executives, which involved more than an announcement of a Defendant’s own conduct. In fact, as alleged below, many of Defendants’ executives’ calls for a new era of “discipline” included explicit statements that deeper production cuts were an industry-wide imperative that would pay dividends for “the industry” as a whole. Defendants, who collectively control nearly 90% of the U.S. chicken market, jointly engaged in this production reduction effort.

I am not sure how I feel about this: It is not obvious where to draw the line between a company that says it is going to cut production because it is going to cut production (and wants its shareholders to know), and a company that says it is going to cut production because it wants its competitors to cut production (and wants to slyly signal that to them). But increasingly this sort of thing seems to be where the action is in antitrust disputes: not secret back-room conspiracies, but (alleged) conspiracies conducted openly but subtly on earnings calls. The more useful securities disclosure is -- the more managers tell their shareholders about the operational details of their companies and their business plans for the future -- the easier it is to concoct and monitor a conspiracy to raise prices. And if all of your shareholders are also shareholders in your competitors -- if they value "the industry" more than the individual companies -- then shouldn't that be what they want?

HFT arms race.

A common complaint about high-frequency trading is that trading firms have to spend more and more money for ever smaller improvements in speed. This speed doesn't really help investors much -- no normal person cares if it takes one millisecond or 10 to execute her stock trade -- but a slow high-frequency trader will constantly be getting picked off by faster ones, so will have to invest just to stay competitive. This seems very wasteful to the HFT firms, who would love to spend less on technology but who have no choice but to keep up with their competitors. It also seems pretty wasteful to many outside observers, who wonder why society expends so much money and brainpower just to make stock trades a few microseconds faster.

I am a bit more optimistic though. If high-speed trading firms want to build state-of-the-art communications infrastructure to connect exchanges in different cities -- if they want to lay fancy cables between New York and Chicago, or shoot lasers across New Jersey -- who am I to complain? It is not obvious that improvements in computer and communications technology will have no benefits for the rest of us, and if excessive stock trading subsidizes infrastructure improvements that otherwise wouldn't get built, that seems like a positive for society.

Anyway here is a delightful story about the UK's National Physical Laboratory and high-frequency trading:

The laboratory is the UK’s national measurement institute — it keeps the official time — and holds responsibility for making thousands of high-precision calibrations for science, physics and technology. Now, it has found a new commercial market: providing a reference point for the superfast computers that power global trading in the 21st century. More than a dozen institutions have signed up, including Intercontinental Exchange, one of the world’s largest exchanges operators, and UBS, the Swiss bank.

“[NPL’s product is] the first thing that is almost creating a network of time around the world,” says Mr Lobo, a former mechanical and optical engineer, who joined NPL in 2011 as head of the group developing atomic clocks and managing the timescale. “We’ve been developing new techniques to disseminate time.” His job now is, in effect, to sell time.

There is a market-structure dimension to this story: As markets operate on ever finer timescales, and as regulators and participants get ever more concerned about fairness and "front-running," it becomes more important to know that, say, my order arrived on Exchange X 10 microseconds before your order arrived on Dark Pool Y, or whatever. At a fine enough timescale that is hard to measure consistently -- different exchanges' clocks can get out of sync -- and so you need a coordinated, and extremely accurate, solution.

But there is also a funding-of-scientific-research dimension to this story: The NPL's work supports basic science, but now it can subsidize that work by selling time to high-frequency traders. It's not cheap to build a cesium clock and synchronize the time around the world with fiber-optic cables, or to do whatever other fiddly measurement stuff the NPL also does. Presumably a lot of that work -- like a lot of scientific research -- is socially valuable but hard to sell. You wouldn't think "time" would be easy to sell either! But high-speed trading makes it lucrative. It does seem like society might benefit from that. 



Deer Park Road Management, the $2.2 billion hedge fund firm founded by Michael Craig-Scheckman, is being investigated by U.S. regulators over whether the hedge fund violated securities laws when valuing infrequently traded bonds, according to people familiar with the matter.

The Securities and Exchange Commission is questioning why the fund priced certain debt positions below market norms, said one of the people, who asked not to be identified because the probe isn’t public. The inquiry focuses on the pricing of hard-to-value residential mortgage-backed securities, some of which can go months without trading.

The more common complaint seems to be that funds overvalue illiquid bonds, which boosts both management fees (the fund manages more assets) and performance fees (they've gone up more), but undervaluing them can work too:

Marking down bonds could lower the value of an entire hedge fund, which would mean less money for fund clients who redeem their investments. The practice could also allow a manager to inappropriately smooth returns or realize a bigger year-end payout.

A fun nefarious hedge-fund strategy would be:

  1. Open a hedge fund with some money of your own.
  2. Raise money from outside investors and buy good illiquid bonds.
  3. Mark those bonds down incorrectly. 
  4. Tell your investors "I am so sorry but we are down 10 percent this year, my bad."
  5. Watch your investors redeem in a huff.
  6. Hand them back their money at the artificially reduced valuations.
  7. Look very sorry to see them go.
  8. Sell the bonds at much higher correct valuations and keep the money for yourself.

I mean, it is not a great idea! (It is certainly not legal advice.) It is more of a "Producers" lucrative-failure strategy for hedge funds. It does not seem to be Deer Park Road's situation, though, since its main fund "returned a cumulative 688 percent from its inception in May 2008 through the end of 2017," and "has never had a losing year."

Blockchain blockchain blockchain.

Here is an email that I got from a reader yesterday:

I thought you would be amused by this distributed ledger some co-workers and I have implemented.

We have a co-worker, Paul, who is notorious for skiving off any chance he gets.  Our boss is out of the office for 2 weeks, so we have a pool going for how many hours he is in the office.

The way we track this is with a distributed ledger.  When (if) the first person sees him in the office they e mail the entire group with just the time and we all enter it into our ledgers.  When someone sees him leaving, they email with the time and we all update it.  Everyone had their own copy making reconciliation easier and without worrying about anyone tweaking the numbers.

8 days in and he is at 10.25 hours.

I mean it is not a blockchain. There are no hashes or digital signatures. But the principle is ... not dissimilar? It reminds me not so much of the bitcoin blockchain -- open, anonymous, permissionless -- as of the private blockchain projects that financial institutions are always announcing. There is a closed universe of participants who all have an interest in the success of the project, and who all trust one another a certain amount: My reader and his work buddies, like JPMorgan and Goldman Sachs, trust each other to generally want to maintain a robust system, even if in individual cases they might want to cheat a bit to favor their own interests. Transactions are broadcast to all participants, and they all keep copies of the ledger. If there are conflicts, you can reconcile them by seeing which version of the ledger has the support of the majority of the nodes in the system. Security is maintained not through cryptography but through distribution: Hacking one person's ledger wouldn't help you because everyone else has an independent copy. It is a nice illustration of how many closed-blockchain-for-financial-institutions projects seem to work, insofar as it has most of the essential features of those projects, and also insofar as it is a bunch of financial-services employees having a bit of a joke. 

The crypto.

This week Facebook Inc. banned ads for "financial products and services that are frequently associated with misleading or deceptive promotional practices, such as binary options, initial coin offerings, or cryptocurrency." That is I think the right way of putting it: There is nothing necessarily fraudulent about ICOs or cryptocurrencies, but they sure do seem to be frequently associated with fraud.

Meanwhile, "Instant buying (and selling, if you don't want to hodl) of Bitcoin is now available to most Cash App customers," tweeted Twitter Inc. Chief Executive Officer Jack Dorsey. (Cash App is a product of Square Inc., the other public company of which Dorsey is CEO, and "hodl" is [sic].) Different approaches!

Elsewhere, here is Nathaniel Popper on worries that Tether might be used to prop up the price of bitcoin. And: "Goldman Sachs Felt Rushed by Quick Rollout of Bitcoin Futures." And: "A Bitcoin Conference Rented a Miami Strip Club—And Regretted It." And: "Aussie man launches own crypto currency, ‘Strayacoin.’" "It’s a parody," he says, "but it’s fully functional." Every cryptocurrency white paper should start with those words. 

The flamethrowers.

I haven't followed Elon Musk's flamethrower-selling stunt closely, but I gather that the gist of it is that he got some flamethrowers, peddled them on the internet for $500 each, and will use the money to fund the operations of The Boring Company, the company he started to (1) drill tunnels or something and (2) conduct hilarious pranks. And it worked: "Elon Musk Stops Taking Flamethrower Orders After $10 Million in Sales," congrats everyone. The Boring Company is not a public company, so I do not expect to follow up in six months with a story about a securities fraud investigation over its failure to disclose to investors that the flamethrowers could light people on fire.

But it is an interesting experiment in corporate finance, no? The proceeds from the flamethrowers "will help bankroll the boring machines, lawyers and lobbyists Musk is relying on to execute his plan to build tunnels under parts of Los Angeles and the Baltimore-Washington Parkway." It is a way to fund the future operations of a startup that:

  1. Is not dilutive (like equity);
  2. Doesn't have to be paid back (like debt);
  3. Doesn't require actually starting the operations (like "bootstrapping"); and
  4. Operates mainly through clever internet hype.

It is not so unlike many initial coin offerings! There are differences, of course: The focus of the hype in this case is a funny (and incendiary) tangible object (the flamethrower), rather than a funny intangible blockchain object (a token). And I doubt that the flamethrower sale is being driven by speculative frenzy, though you never know; the Boring flamethrowers are a numbered limited edition and you could imagine them becoming collectors' items. Still, a basic lesson of the ICO boom is that people will happily fund companies in non-traditional ways that are strictly worse for investors than the traditional ways, as long as they seem cool on the internet. That lesson might be relevant beyond ICOs themselves. I look forward to other companies raising money by selling branded laser guns and jetpacks and hoverboards.

Things happen.

Hedge Funds Just Had Their Best Year Since 2010. Steve Cohen Returns With ‘Best Ideas’ Trading Book. Steve Cohen Needs to Prove Himself to Wall Street All Over Again. How GE Went From American Icon to Astonishing Mess. EU seeks powers to stop post-Brexit bonfire of regulation. EU rejects Brexit trade deal for UK financial services sector. Ex-Morgan Stanley Brokers Admit to Using Client Cash for Wind Farm. High-Interest Lenders Benefit From Trump’s Deregulatory Stance.What's Left Shark up to? The Awl, 2009-2018. Michigan dog approved for unemployment benefits

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.