Streetwise Professor

May 1, 2018

Cuckoo for Cocoa Puffs: Round Up the Usual Suspects

Filed under: Commodities,Derivatives,Economics,Exchanges — The Professor @ 10:39 am

Journalism on financial markets generally, and commodity markets in particular, often resorts to rounding up the usual suspects to explain anomalous price movements.  Nowadays, the usual suspect in commodity markets is computerized/algorithmic/high frequency trading.  For example, some time back HFT was blamed for higher volatility in the cattle market, even though such trading represents a smaller fraction of cattle trading than it does for other contracts, and especially since there is precious little in the way of a theoretical argument that would support such a connection.

Another case in point: a flipping of the relationship between London and New York cocoa prices is being blamed on computerized traders.

Computers are dominating the trading of cocoa in New York, sparking a dramatic divergence in the longstanding price relationship with the London market.

Speculative funds have driven the price of the commodity in New York up more than 50 per cent since the start of the year to just under $3,000 a tonne. The New York market, traded in dollars, has traditionally been the preferred market for financial players such as hedge funds.

The London market, historically favoured by traders and commercial players buying and selling physical cocoa, has only risen 34 per cent in the same timeframe.

The big shift triggered by the New York buying is that its benchmark, which normally trades at a discount to London, now sits at a record premium.

So, is the NY premium unjustified by physical market price relationships?  If so, that would be like hundred dollar bills lying on the sidewalk–and someone would pick them up, right?

Not according to this article:

The pronounced shift in price relationships comes as hedge fund managers with physical trading capabilities and merchant traders have exited the cocoa market.

In the past, such a large price difference would have encouraged a trader to buy physical cocoa in London and send it to New York, hence narrowing the relationship. However, current price movements reflected the absence of such players, said brokers.

Fewer does not mean zero.  Cargill, or Olam, or Barry Callebaut or Ecom and a handful of other traders certainly have the ability to execute a simple physical arb if one existed.  Indeed, given the recent trying times in physical commodity trading, such firms would be ravenous to exploit such opportunities.

What’s even more bizarre is that pairs/spread/convergence trading is about the most vanilla (not chocolate!) type of algorithmic trade there is, and indeed, has long been a staple of algorithmic firms that trade only paper.  Meaning that if the spread between this pair of closely related contracts was out of line, if physical traders didn’t bring it back into line, it would be the computerized traders who would.  Yes, there are some complexities here–different delivery locations, different currencies, different deliverable growths with different price differentials, different clearinghouses–but those are exactly the kinds of things that are amenable to systematic–and computerized–analysis.

Weirdly, the article recognizes this

Others use algorithms that exploit the shifts in price relationships between different markets or separate contracts of the same commodity. [Emphasis added.  I should mention that cocoa is one of the few examples of a commodity with separate active contracts for the same commodity.]

It then fails to grasp the implications of this.

One “authority” cited in the article is–get this–Anthony Ward of Armajaro infamy:

Anthony Ward, the commodities trader known in the cocoa market for his large bets, has been among the more well-known fund managers to close his hedge fund, exiting the market at the end of last year. Mr Ward, dubbed “Chocfinger” due to his influence over the cocoa price, blamed the rising power of algorithmic and systems-based trading for making position-taking based on “fundamental” supply and demand factors more difficult.

Methinks that the market isn’t treating Anthony well, and like many losing traders, can’t take the blame himself so he’s looking for a scapegoat. (I note that Ward sold out Armajaro’s cocoa trading business to Ecom for the grand sum of $1 in December, 2013.)

I am skeptical enough that computerized trading can distort flat prices, but those arguments are harder to refute because of the knowledge problem: the whole reason markets exist is that no one knows the “right” price, hence disagreements are inevitable.  But when it comes to something as basic as an intracommodity spread, I find allegations of computer-driven distortions completely implausible.  You can’t arb flat price distortions, but you can arb distorted spreads, and that business is the bread and butter for commodity traders.

So: release the suspect!

PS. For my Geneva students looking for a topic for a class paper, this would be ideal. Perform an analysis to explain the flipping of the spread.

December 26, 2010

The Last Shovelful of Dirt

Filed under: Commodities,Derivatives,Economics,Exchanges — The Professor @ 4:00 pm

No doubt when you were quaffing down a warm cup of hot chocolate on Christmas, you were thinking to yourself: “I wonder how that cocoa corner played out.”  Well, maybe not.

But anyways, wonder no more.  In mid-December, Armajaro, the British hedge fund run by Tony Ward, delivered the bulk of the 110,000 tons of cocoa tendered against the expiring contract.

The deliveries occurred at a price of about 2000 BP/tonne; Armajaro took deliveries at prices of around 2700 BP/tonne. On deliveries of about 240,000 tons, that corresponds to a loss of 168 million BP, or about $260 million.

That, boys and girls, is what they call “burying the corpse.”*

Armajaro’s losses on the deliveries it took might actually exceed this outsized sum if they made any deliveries or cash sales at spot prices prior to December.  Prices were below 1800 BP/tonne in November; only tragic political turmoil in the Ivory Coast has buoyed prices of late.  But even chaos in the most important cocoa producer (something not anticipated in July) has not been enough to drive prices back to the inflated levels seen in July.

In his recent article on the cocoa market, FT commodities editor Javiar Blas duly notes that Armajaro delivered a huge quantity of cocoa.  But he lets pass in deafening silence the fact that this large delivery completely undercuts his previous “reporting” on Armajaro’s actions.  In this case, “reporting” meaning “credulously repeating Tony Ward’s fantastical cover story.”  A cover story, I might add, which was transparently fantastical when it was first spun in July–as I pointed out then.  You might remember the bologna about expectations of a bad crop (which would in no way make it rational to take huge deliveries in the face of a steep backwardation) and pre-sales of the deliveries (which would have, in fact, enhanced the profitability of a manipulative strategy, and the existence of which, in any event, is belied by December’s deliveries). Self-evident and self-serving tripe then, and now.

Blas writes:

In any case, investors will do well not to focus on Armajaro’s selling, but on the buying side. If big physical players turn to the exchange for supplies over the short term, it could be a signal that the cocoa market is heating up again.

Yeah, if I were him I wouldn’t want anybody focusing on Armajaro’s selling either.  Sorry, but “move on, nothing to see here” doesn’t quite cut it.

Armajaro’s escapade sparked considerable popular pique .  I had quite a few people contact me to register their outrage at Ward’s machinations.  Someone started an anti-Armajaro Facebook group, and a couple of British artists also tried to publicize the shenanigans.

The exchange and the British regulators–not so much.  And the FT–definitely not so much at all.  Quite the contrary.  It’s sad commentary when some ordinary folks on Facebook have better BS detectors than those operating the market, those supposedly overseeing it, and those who get paid to report on it.

* And no, Mr. Nissen, I didn’t coin that phrase.  But I wish I had.

October 2, 2010

Tony Ward’s BS Exposed by the FT–But the FT Doesn’t Even Know It

Filed under: Commodities,Derivatives,Economics,Exchanges — The Professor @ 12:31 pm

Weather in Ghana has been good to the cocoa crop this year, which is putting downward pressure on prices:

After four years of shortages that sent the price of cocoa to its highest level in three decades, traders almost unanimously believe that the cocoa market is set to move into a surplus in the 2010-11 marketing year, which starts on Friday.

The looming oversupply has sent cocoa prices down by a third in the past two-and-a-half months.

The surge in production in west Africa, which accounts for nearly 70 per cent of the world’s supplies of the commodity used in chocolate, is a setback for investors such as hedge funds Armajaro and Clive Capital, which bet on rising prices earlier this year.

The cost of cocoa in London – the benchmark – has fallen to about £1,900 a tonne, down from a 33-year high of £2,732 a tonne set in July.

So Tony Ward must be hurting, right?  After all, he told the world–the world–that due to his super-sophisticated weather forecasting system, a proprietary network of weather stations, and the best weather data in the world, he knew a crop shortage was coming. So he must have been way long, and the price fall must have crushed him.

Well no, actually:

Investors say that Anthony Ward, the founder of Armajaro’s flagship CC+ fund, has suffered less than expected, however, as bearish hedges cushioned the fund from the price drop.

Oh, really?  Hedges, huh?

This means that the story he was flogging in July and August was complete bilge.  He wasn’t bullish on the new crop. It would be more accurate to say that he was bullsh*t on the new crop.

Which also means that his cover story for taking delivery in July was also complete bilge/bullsh*t.

Not that I am the least bit surprised.  The story made no sense anyways: it’s nuts to take delivery of the old crop at a big backwardation just because you think the new crop price is going to shoot up.  If you think that, buy the new crop, not the old crop. Duh.

Here’s a far more credible explanation of what happened.  Armajaro cornered the cocoa market.  Being clever, Ward knew the most difficult part of cornering is burying the corpse, so he did that by obtaining a burial plot in advance: the “bearish hedges” mentioned in the article.  All of the new crop shortage stories were just so much eyewash to fool the gullible.

Of whom, as this article and others indicate, FT reporters–most notably Javier Blas–are the foremost.  If they won’t get a room, could they at least get a clue?

July 24, 2010

Get a Room

Filed under: Commodities,Derivatives,Economics,Exchanges — The Professor @ 10:17 pm

Today’s FT runs another cringe-worthy PDA with Anthony Ward of Armajaro.  It is as probing and hard hitting as anything you would expect to read in, say, TigerBeat.  Its tone runs the gamut from credulous to worshipful.

I found the last two paragraphs particularly entertaining:

He is said to argue now as he did in 2002, that all he has done is go to the futures market “to buy cocoa in the most efficient and low-risk way possible”. The trade is likely to bring a stream of profits to Armajaro, particularly if the new Ivorian crop, due in October, disappoints.

Even if prices decline, Mr Ward is unlikely to suffer. A competitor who knows him well has no doubt he will have hedged his position. One admiring executive says: “No one knows the cocoa market better than Anthony.”

Paragraph 1: He’s long cocoa, so if prices rise due to a disappointing crop he reaps a stream of profits.

Paragraph 2: He’s hedged, so if prices fall, he doesn’t lose.  That is, he isn’t long cocoa.

If both of those things are true, he would indeed be a great trader: prices rise he wins, prices fall, he doesn’t lose.  He’s long the upside but not the downside.

How does that work, exactly? I mean, these diametrically opposed statements are in adjacent sentences.  Didn’t an editor think that such a feat was unlikely, or at least sufficiently intriguing to demand further investigation, explanation, and reconciliation?

There is, of course, a way that it could work: Armajaro could be long cocoa puts.  But this seems highly unlikely, as the traded put market is small relative to the deliveries that Ward has taken.  At the very least, though, the tension between the last two paragraphs should have spurred the report to do some actual, you know, investigation and reporting, to see whether the put story has any factual basis, and if so to provide some background on the cost of the puts and its effect on the profitability of the strategy.  And if it doesn’t (as is likely the case), to state forthrightly that Armajaro’s position is a risky one subject to a risk of losses symmetric to the prospect of gains.

But instead the reader gets a breathless fan mag portrayal of the hero who is able to do things that are impossible, or at least so unlikely as to demand further explanation.

This is all bad enough, but nothing compared to the apparent failure to ask any aggressive questions to test the reasonableness of Ward’s denial of squeezing or cornering the market.   If the questions were asked, but not answered, that should have been made plain, or the piece shouldn’t have run at all.  But that would ruin a nice piece of hagiography.

Mr. Ward’s evident hold over the FT is very curious.  An enterprising competitor could do worse than trying to figure out why–after, of course, doing some serious reporting on what has transpired in the cocoa market in recent weeks.

July 22, 2010

Chocolate Kisses, or the FT in the Tank

Filed under: Commodities,Derivatives,Economics,Exchanges — The Professor @ 10:34 am

I have been a close observer of manipulation since the mid-1980s, when the Japanese were squeezing the 30 year while I was devising interest rate hedging strategies for the clients of the FCM I worked for.  Since then, I have examined, intensively and forensically, manipulations in soybeans, Brent, Bunds, Treasuries, canola, copper, aluminum, propane, natural gas, WTI, and other things.  I have researched every major manipulation that I could identify, dating back to the 1860s.

I have heard all of the lame alibis, sometimes directly from the manipulators’ mouths, sometimes from the mouths of their mouthpieces.  Excuses from the likes of them are to be expected.  It is disappointing, however, to see the Financial Times regurgitate their tripe.

Yesterday the FT ran an editorial that recycled many of the tired tales that manipulators tell in defense of Anthony Ward’s actions in the cocoa market.  None of it is exculpatory, in the least.

To start with, the FT opines:

Armajaro breached no rules when, in pursuit of its view that cocoa prices would rise after a poor harvest, it snapped up contracts to purchase cocoa beans on the Liffe futures exchange. These gave it the right to take physical delivery of the beans if it chose. What stirred the controversy was that Mr Ward exercised that prerogative over 24,100 contracts. These amounted to about 7 per cent of annual world production.

Yes, the old “I had the contractual right to do what I did” defense.  But a market power manipulation involves the opportunistic use of this contractual right to distort prices and the flows of a commodity in commerce.  This action degrades the effectiveness of futures contracts as a risk shifting and price discovery mechanism, and leads to wasteful use of the commodity.  If “contractual right” trumps everything, then rules against corners are a nullity.  If that’s the way the game is played, maybe I should start The Corner Fund and go to town.  I’ll hire the FT to flack for me.  But I probably won’t even have to hire it: it seems it’s willing to do it for free!

The FT continues with another old standby: the “They had it coming to them for walking in this neighborhood alone dressed in shorts like that” story:

There is no doubt that his decision caused agony for some on the other side of the contracts. A number of cocoa traders use Liffe to hedge against falls in the prices of the stocks they hold. They suffered because they not only hedged their beans – but their holdings of cocoa powder and butter also. These were ineligible for delivery against the contracts, thus forcing the hedgers to settle contracts for cash at penal premiums or to buy sufficient beans to close the trade.

It is hard to see how Mr Ward can be blamed for this. The hedgers did not need to run a mismatch risk. They could have found other ways to insure their exposures – such as entering into a derivative with a bank. That would have been more expensive, but would also have been safer.

But the whole point of centralized markets is to concentrate liquidity, thereby permitting out-of-position hedgers to reduce risk with maximum flexibility while incurring low transactions costs.  They willingly trade basis risk for transactions costs.  That is no reason to excuse opportunistic actions that exacerbate basis risk.  The risks of being cornered induce hedgers to take costly preventative measures, such as investing collecting information about the activities of other traders that is socially wasteful because it is merely intended to reduce the risk of being the source of a wealth transfer, or using markets where transactions costs are higher.

Beating me to the punch, Tullet Prebon’s CEO T.C. Smith skewers the FT’s fatuousness–and hypocrisy:

Many commentators maintain that one lesson of the financial crisis is that all over the counter derivatives should be traded on exchanges.

I wonder how the revelation in your editorial, that the cocoa consumers who attempted to hedge their powder and butter through the imperfect mechanism of the Liffe futures in cocoa beans would have suffered much less if they had taken out OTC contracts with banks which precisely hedged their risk, fits their theory.

They can file your editorial with the reports on the as yet unexplained “Flash Crash” of May 6, in which Apple’s shares traded at one cent per share and $100,000 a share within 20 minutes.

But then, why let the facts get in the way of their theory? It will all be safer and more transparent when everything is on exchange.

Couldn’t have said it better myself.

The FT then plays three card monte and attempts to distract attention from the real issue:

It has been suggested that Mr Ward’s conduct was akin to cornering, and thus led to a disorderly market. But this is a hard case to sustain. Disorderly markets occur when two-way prices cease to be made – thus making cash settlements impossible. But the open interest on the cocoa contract fell in the weeks prior to expiry.

That is, shall we say, an idiosyncratic definition of “disorderly market.”  In fact, in many corners trading activity can reach a frenzy.  There are two-way prices, but the prices are far above where they would be in a competitive market, and the “cash settlements” are at essentially extortionary prices.  Apparently that’s hunky dory with the FT.  And even in cornered contracts open interest typically falls.  In fact, that’s almost always true.  Indeed, it often falls the most when the cornerer makes money by liquidating futures at supercompetitive prices.

The FT then suggests that the quiet word among gentlemen approach is preferable in such matters:

However, squeezes can be managed. Liffe tries to do this by directly stepping in where necessary to guide the activities of traders. This approach, which has been compared to being summoned to the “headmaster’s study for a quiet word”, effectively substitutes for a detailed rule book. The flexibility it provides can be valuable so long as the market is actively policed.

Yeah.  That works, except when it doesn’t.  And the London markets provide numerous examples of it not working.  Remember Sumitomo, and the LME’s rather, shall we say, prone, not to say accommodating, posture in dealing with Hamanaka right to the end?  Or all of the Brent shenanigans that occurred over the years?

And in that vein, a shout out to LIFFE, for validating my 1995 JLE article arguing that exchanges were unlikely to act against manipulation.  It would be hard to write a letter that does a better job at missing the elephant in the room.

Perhaps due to space limitations, the FT does manage to overlook some of the standard lines from the manipulator’s manual.  But a couple of commentors over at SeekingAlpha fill the void.

In this case, it appears that Armajaro already had customers for the cocoa delivered to it.  See!, the commentor says, there is real demand, real customers.  Well, if you know anything about manipulation, you know that the biggest danger is related to burying the corpse of the manipulation: that is, selling the massive deliveries that you take to squeeze the market.  A clever fellow bent on manipulating the market pre-arranges the funeral.  If the reports about forward sales of old crop cocoa to processors are correct, that was done in this case.  So, rather than being exculpatory, such sales are actually evidence of manipulative intent.

(I’d be interested to know who is making the deliveries.  It would be telling if any of the deliveries were made by those who had contracted to buy from Armajaro.  It would also be interesting to know more of the details of those contracts, not just the pricing terms but any other features in the contracts relating to use or marketing of the cocoa, and the delivery locations for the contracts.)

The other common excuse, again seen in my SA comments, is that the stuff that is delivered will be consumed.  Well, duh.  It’s not like the guy is going to eat it all himself, or burn it, or build houses out of cocoa beans.  It is going to be consumed.  But manipulation distorts consumption pattens–the timing and location of consumption.  The stuff is consumed, but in the wrong place at the wrong time, and too much of it is shipped around to the wrong places.  The fact that the delivered cocoa will eventually be consumed does not imply that the deliveries are efficient.

All in all, none of the arguments raised in Armajaro’s defense are even remotely exculpatory, and some are actually adverse.  A final judgment would require a full forensic and econometric evaluation of prices, pricing relationships, price-quantity relationships, and movements of cocoa.  Time permitting I hope to do some of that.  But at this juncture, there is sufficient evidence to conclude that there is a colorable case against Armajaro, and that the arguments raised in its defense are risible.

Taking huge deliveries in a large backwardation is consistent with manipulation.  This is especially true when one party takes all the deliveries.  Delivery economics are pretty straightforward, and if they work for one party they tend to work for several.  One firm taking all of the deliveries is suspect.

Moreover, the purported rationale for the trade–an expectation of a small new crop in the autumn–does not explain why the old crop price richened relative to the new crop price.  If the real motivation was a view that the market did not appreciate sufficiently the likelihood of a small crop, then the trade should have been to go long the new crop and cash in if and when that view was validated.  Taking delivery into a big backwardation, and seeing the backwardation actually increase as result of the deliveries, is not a sensible way to play an anticipated shortage in the new crop.  The backwardation across crop years screams: “don’t carry over inventory into the new crop year.”  So current inventories do nothing to alleviate future shortages (because the spreads signal that all of the cocoa should be consumed before those shortages arise).  (As an example of how the market should work in response to anticipation of a crop shortfall, during the big US drought of 1988, the old crop-new crop spreads for corn and soybeans snapped into a huge carry; indeed, the spreads exceeded full carry charges measured using official storage rates.  The warehouses in Chicago were filled to the top with grain and beans. That’s because the market realized that it was desirable to carry over old crop supplies in larger quantities than usual to alleviate the anticipated shortfall in the 1988-1989 crop.)  (Ferruzzi told the same lame betting on a new crop shortfall story during its bean manipulation in 1989.)

The rise in the London price relative to the NY price is also what you’d expect to observe in a corner.

I’m not rushing to judgment.  But neither should the FT .  Its unseemly haste to play Ward’s mouthpiece, its heaving up of tired excuses made for manipulators from time immemorial, and its complete lack of any skepticism, are not befitting a premier financial publication.

Powered by WordPress