Crop price anomalies baffle economists

March 30, 2008 2:22 PM

Odd Crop Prices Defy Economics.For finance and economics geeks: Could a drugstore sell two identical tubes of toothpaste, and charge 50 cents more for one of them? Of course not. But, in effect, exactly that has been happening, repeatedly and mysteriously, in trading that sets prices for corn, soybeans and wheat — three of America’s biggest crops and, lately, popular targets for investors pouring into the volatile commodities market. The curious thing is that these price anomalies should be ripe for arbitrage. There should be no gap between the price of say, wheat in the cash market and the wheat futures contract on the day the contract expires.
posted by storybored (25 comments total)

Look, what I want is financial instruments, not some fucking plant that grows in a flyover state, capisce?
posted by fleetmouse at 3:01 PM on March 30, 2008 [2 favorites]

Could we back a fiat currency with this stuff?
posted by bonaldi at 3:07 PM on March 30, 2008

Gah, the opposite of fiat, of course. shit.
posted by bonaldi at 3:08 PM on March 30, 2008

Yeah, wtf is up with the arbitrage traders? What do they know that we don't?
posted by agentofselection at 3:18 PM on March 30, 2008

Economists: "we have no idea what's going on" - awesome
posted by stbalbach at 3:26 PM on March 30, 2008 [1 favorite]

Hmm, didn't I hear this characterized as a mad dash from the collapsing hedge funds (or any other funds backed by mortgages) into anything that would hold value, thus giving commodity markets unusual volatility? Reading a lot of economics blogs these days...maybe I can find a link.
posted by telstar at 3:36 PM on March 30, 2008

Hmm, it's a little weird but I don't quite see why they're so concerned.

As futures contracts get close to expiration, traders roll them over into later contracts, mainly because you don't want to get caught having to deliver, or worse, having to accept delivery of a boxcar of wheat somewhere.

So by the time it expires, there simply aren't that many of them out there, so the market simply isn't that liquid.

I don't see a really good way to efficiently enforce this arbitrage without getting into the wheat reselling business, and there probably isn't enough money to made in it.
posted by lupus_yonderboy at 3:40 PM on March 30, 2008 [1 favorite]

Oh, wait...you mean markets don't operate exactly the way theory says they should? That this wildcard called "people" might make things act illogically? How interesting...
posted by Thorzdad at 3:45 PM on March 30, 2008 [1 favorite]

Some people in the article speculate this is the result of speculators who have no idea what they're doing. I would guess the same thing.
posted by delmoi at 4:02 PM on March 30, 2008 [1 favorite]

Wow. That was an amazingly uninformative article. Poor explanation of futures contracts, no explanation of what the "expiration" price of a futures contract actually is, no indication of the volumes trading at different prices simultaneously, no indication of who is paying too much (or accepting too little), poor indication of magnitudes or frequencies of these anomalies, etc. Basically, the article is a big press release for some economists who published a paper. The only thing I can see that the article gets right is in linking to the actual paper, which itself is not that illuminating. I went through a Ph.D. program in economics, and I still don't understand why I should care about these results, or what they mean. I take it the delivery obligations under futures contracts are traded - so that one might expect the trading prices to behave like those of zero-coupon bonds: as maturity (or expiration of the contract) approaches, the trading price should approach the face value (or spot ("cash") price, for commodities), since there is less time between the time of the trade and the time of delivery for new information to affect traders' willingness to pay (or accept) a particular price for the commodity. But I still don't understand: are people actually trading futures contracts the day before delivery is due on them at substantially different prices than the next day's spot price? Are these prices different by more than one would expect when assuming someone else's contract, that may include sub-optimal delivery, risk-of-loss, or quality terms? Does anyone actually understand what this article is about?
posted by dilettanti at 4:04 PM on March 30, 2008

telstar I heard a similar piece on Marketplace a month or two ago, damned if I can find a link though.
posted by Skorgu at 4:14 PM on March 30, 2008

There is actually a pretty straightforward answer; the futures contract is priced at a premium because of the time between expiration and delivery. If you think the underlying is going up nonstop (which they are), put a couple price on the spread.

For example, take the March Corn contracts on the CBOT. Expiration date is 3/14/2008, last delivery is 3/18/2008. (see: http://www.cbot.com/cbot/pub/page/0,3181,1082+C+3+2008,00.html).

Now with corn trading at 560, a 10c premium for 4 calendar (2 trading) days of risk is actually amazingly low. Put another way, when you are long a contract into expiration you actually get the underlying + a 4 day call option. The premium is the price of the call option. (If corn goes up $100/bushel from the 14th to the 18th you still buy it for the 14th price but you could sell it for the 18th days price at a nice profit...)

I have seen a bunch of people scratching their heads about this stuff - to me the story is watching all these non-market people trying to understand the mechanics of the market. Its cool, but places like the NY times and posters like above who think this is all a big mystery are doing a disservice.
posted by H. Roark at 5:36 PM on March 30, 2008 [3 favorites]

To put it more clearly, this snippet from the article is just wrong:
And on expiration day, when the bushels of wheat covered by that futures contract are due for delivery, their price should very nearly match the price in the cash market, allowing for a little market friction or major delivery disruptions like Hurricane Katrina.

expiration is not equal to delivery.
posted by H. Roark at 5:41 PM on March 30, 2008

Could we back a fiat currency with this stuff?

Fiat currency is by definition backed only by hot air and threat of law.

The notion of an agricultural commodity backed currency, however, has been mooted by Frank M. Bryan and Bill Mares in their novel about Vermont's hoped for secession. They chose maple syrup, since it is hard to make, hard to fake, widely treasured, and leaves a sticky trail if one tries to steal it.
posted by IndigoJones at 6:06 PM on March 30, 2008

Wow. That was an amazingly uninformative article.

I have to agree with you on that - it would have been nice if the author had dug a bit more deeply and described what actually happens on the expiration date, the mechanisms involved in closing the contracts and so on.
posted by storybored at 6:07 PM on March 30, 2008

H.Roark: expiration is not equal to delivery.

So now I'm even more confused, and more disappointed in the article. The only concrete piece of information in the article was wrong? What does the "expiration" date for a futures contract signify? The paper that inspired the article in the FPP suggests that we should expect convergence of expiration and cash prices - but if there are several days of trading in between (especially considering the increased volatility the paper also documents), then why is it alarming that there is a difference in these prices?

H. Roark: posters like above who think this is all a big mystery are doing a disservice.

I'm not entirely sure you're referring to me, since I thought the article was mysterious, not the results the economists reported, but just in case: I certainly didn't mean to do anyone a disservice by asking for clarification on a poorly written article - I apologize for any inconvenience my confusion might have caused. I take it the answers to my initial questions are: no, no, and yes (just not me)?
posted by dilettanti at 8:50 PM on March 30, 2008

I was referring to thorzdad, with his/her whole "wow people in the market are idiots to think their models dont work when people are involved" attitude. Quant finance has enough of an image problem already...I guess i'm just a bit sensitive to it.

As for what "expiration" means, look at the link I posted above. Note there is no date for "expiration" - there is a final trading date, a final delivery date etc. If they are talking about market prices on "expiration" not converging they must be talking about the last trading date - which its obvious is different from the final delivery date.

For comparison, look at this: SP500 Calendar - this is the calendar for SP500 futures which are cash settled (as opposed to physical delivery for most agricultural contracts). Note that with cash settlement, the last trading day is equal to the settle date. Here, the notion of "expiration" makes more sense - the value of the futures contract should equal the index value and settlement happens at the same time.



I think the author (and by extension their economists) dont understand the mechanics of the futures market.
posted by H. Roark at 9:28 PM on March 30, 2008

It looks like straightforward market manipulation by really big players to me.

There's only a certain amount of wheat or corn (say) being grown in the world at a given time or in storage, and the futures contracts seem to last a maximum of three months, so no matter how high futures contracts are bid up, the amount of product which can appear for sale is very limited. This means that you can know in advance to a reasonable certainty, by estimating the available amounts of the commodity in the world, just how much money it will take, pumped into futures markets, to make the price of the commodity rise sharply.

A big player, or a syndicate of big players seeing this, takes a huge position in the futures market, which drives up the price of the commodity tremendously because there's not that much left for everyone else to buy-- and people will insist on imagining that they should have something to eat when they are hungry, so he can count on people continuing to buy the commodity no matter how high the price. Then the BP holds on to his position as long as he can as the expiration date approaches, and dumps his holdings at the last minute to take advantage of the high prices his big position has created in the first place.

The discrepancy between the cash price and the futures price happens when the prices of the futures contracts and the commodity start to fall so fast as he sells off his position the big player concludes that if he sold all of his futures contracts by the expiration date there would be so much available in such a short time that demand would be overwhelmed and no buyers would appear, the price would go into free fall, and he would lose more money than if he just held them and ate the losses, which are more than covered by gains on the contracts he was able to sell, anyway.

The only problem is that such discrepancies are dead giveaways to regulators, or would have been to the regulators of any previous Administration, at least.
posted by jamjam at 9:33 PM on March 30, 2008

H. Roark: What's the delivery time for a cash contract then?

I'd have assumed that both the cash and future contracts would have the same delivery restrictions, once trading is complete. You seem to be suggesting that's not so. Am I missing something?
posted by mr. strange at 10:10 PM on March 30, 2008

jamjam: There are a few ways to test your theory with the actual data. Why don't you do so?
posted by ssg at 10:34 PM on March 30, 2008

Then the BP holds on to his position as long as he can as the expiration date approaches, and dumps his holdings at the last minute to take advantage of the high prices his big position has created in the first place.

Oh, why didn't you just say so, then.

It's like Trading Places.
posted by Civil_Disobedient at 3:39 AM on March 31, 2008 [1 favorite]

strange - There are usually only one type (cash / physical) type of contract for an underlying. For example, there are no physical settlement contracts for the SP500 indexes (think about the fractional share nightmare etc).

If there are two type of contracts, then yeah they would be priced differently.

Another example that I bet would cause similar confusion is that the current values of the SP500 contract do not match the values of the indexes. The easiest way to think about this is the future does not receive dividends while holding the underlying (or SPY etf) does.
posted by H. Roark at 5:54 AM on March 31, 2008

Part of the problem with the corn futures is the government subsidizies for E85. It's cheaper at the pump since the government is covering 50 cents of it. Costs more to produce, we use corn which drives up the price for groceries and also feed for cows and chickens so their cost goes up.

I think the economists could do just as well gazing into a crystal ball.
posted by tabberone at 6:07 AM on March 31, 2008

Maybe there's a level of irrationality the economists aren't accounting for?
posted by drezdn at 8:32 AM on March 31, 2008

Semi-related, today from the NYT: Grains Gone Wild.
posted by salvia at 6:52 AM on April 7, 2008