Comments on: How commodity indices broke the wheat futures market A slice of lime in the soda Sun, 26 Oct 2014 19:05:02 +0000 hourly 1 By: dWj Fri, 26 Jun 2009 13:09:32 +0000 The existence of the delivery oligopoly is necessary and nearly sufficient. That should be considered the primary cause of the failure here, even if it needed a trigger before the spreads got particularly wide in the last couple years. Making the futures cash-settled requires a way to determine the authoritative final settlement price; if it’s not too messy (in terms of performance risk) to dramatically expand the number of permitted deliverers, that would seem to me to be the sensible solution.

By: csissoko Thu, 25 Jun 2009 17:28:32 +0000 To be more precise a historian can compensate for underlying events such as changes in the legal environment, in norms in conduct of markets, in the number and quality of market participants, etc. In short the world is full of important, but unmeasurable variables.

By: csissoko Thu, 25 Jun 2009 17:18:25 +0000 Nick, My view is that while the human brain can incorporate very complex and subtle factors that make historical analysis useful when used with great caution, the idea that you can turn this extremely subtle understanding of events into a series of data points strikes me as utopian.

By: Nick Thu, 25 Jun 2009 16:32:54 +0000 Csissoko–if one claims that the event of interest was so “time and place specific” that one rejects the possibility of using historical data to analyze factors at work, logical consistency would demand that one should also not use the analysis of the event in question as an input to policy decisions about the future, which is a different time and place than the event in question.

By: csissoko Thu, 25 Jun 2009 15:37:30 +0000 Matthew: Maybe the explanation is a combination of Nick’s and the report’s — there was hoarding and futures selling, but the average farmer was prevented from profiting because when storage reached its limits, the delivery oligopoly allowed the basis spread to go astronomical. (I know nothing about the market so this is pure speculation.)

Nick: “one of the problems with this report is that it doesn’t even attempt to test its hypothesis against the data in a statistical framework” This approach works in many cases, but some interesting models in the social sciences depend on unmeasurable variables like expectations. While using prices calculated from derivative contracts can proxy for expectations, they also introduce “technical factors” (i.e. market aberrations) with hard to assess biases.

Furthermore, it is important to understand that data analysis is extremely limited in what it can tell a social scientist, because many of the interesting questions involve counterfactuals — and the data to test them does not exist — unless you are willing to make strong assumptions about the event you’re interested in not being time and place specific (and therefore being sufficiently comparable to use data from “similar” events to analyze the event you’re interested in).

Do you really want to take the position that we should discard out of hand all models that rely on unmeasurable inputs? Personally I think that would be foolish in the extreme.

By: Matthew Thu, 25 Jun 2009 14:00:19 +0000 I think the important paragraph is this:

“higher futures prices made it more profitable for grain elevator operators to purchase grain in the cash market, place it into storage, and then hedge those grain purchases with the sale of relatively high-priced futures contracts than to engage in arbitrage transactions (buying wheat in the cash market, selling futures contracts, and then delivering the wheat) at contract expiration”

But I don’t get it. The actions are the same initially – the elevator operators are buying (cheaper) grain in the spot market and selling forward (more expensive) grain in the futures. Then, however, the report says that they are holding onto the grain (so presumably settling the sold futures in cash) rather than delivering into them. But – and surely this is the nub of the entire argument – why are they doing this? It doesn’t seem profitable.

By: Nick Thu, 25 Jun 2009 12:49:23 +0000 Agreed with Matthew–Csissoko’s (and the report’s) argument that higher futures prices could lead to higher expected future spot prices should mean that current spot prices go *up* if anything, since wheat is a storable commodity. The report is suggesting that there is a positive basis because holders of physical wheat are hoarding it, but if that is indeed the case, then the quoted spot price must not really be the market price because suppliers are unwilling to sell at that price.

On the discussions about econometric analysis, of course I don’t think that a regression alone can provide a conclusive answer to the question. A thorough investigation would use both qualitative analysis (such as interviews with market participants) and quantitative analysis (such as econometric modeling). While a model being consistent with the data does not prove that the model is correct, a model being inconsistent with the data is generally taken as strong evidence that the model is incorrect. This is exactly why statistical analysis uses the terminology “reject the null hypothesis” or “fail to reject the null hypothesis” rather than “accept the null hypothesis.” And one of the problems with this report is that it doesn’t even attempt to test its hypothesis against the data in a statistical framework.

I’d also like to expand on the point that the basis may be due to the oligopoly on physical delivery granted by the exchanges. Let’s see who holds the four delivery licenses, according to the report: ADM, Cargill, Nidera, The Andersons–large agribusiness companies that generally buy grain from farmers. If they are hedging in the futures market, they would be buyers of futures contracts. If futures contracts are closing above the spot price, that means Cargill et al are making money, so what incentive do they have to arbitrage the basis away through physical delivery? I’m not saying that my speculation is conclusive proof of this argument, just that the issue is worth thinking about. A positive basis at contract expiry is a risk-free arbitrage, so the fact that no one is executing it suggests that there are some barriers to doing so. And again, if what Congress really cared about was making the spot and futures prices converge at expiry, all they would have to do would be to require the contract to be cash-settled using the spot price as the reference price.

By: Matthew Thu, 25 Jun 2009 07:30:07 +0000 Sorry, penultimate sentence should read ‘Indeed why AREN’T arbitrageurs buying spot wheat…’

By: Matthew Thu, 25 Jun 2009 07:29:00 +0000 I’m not sure I follow its certainty either. It says:

In a properly functioning futures market, futures and cash prices converge as futures contracts near expiration. Otherwise, if one price were higher, a trader could buy he commodity in the lesser-priced market and immediately sell it in the higher-priced
market for a quick profit. Those types of transactions would soon equalize the two prices. But on many occasions during the last few years in the Chicago wheat market, the two prices have not converged.

One key reason is that the large price disparity between the cash and futures price makes it much more profitable for grain merchants to buy grain in the cash market, hold onto it, and then sell it later—at the price of the higher-priced futures contracts—than engage in the type of transactions described above between the cash and futures market that would make the two prices converge. In addition, the large price disparity means
hat merchants who already have grain in storage and have hedged that grain by selling futures contracts could suffer a loss if they decided to actually sell their grain in the cash market, because they also would have to buy back the futures contract at a higher price
han they could get for selling their grain in the cash market.

but why doesn’t “grain merchants to buy grain in the cash market, hold onto it, and then sell it later—at the price of the higher-priced futures contracts” increase the spot price relative to the futures price? Indeed why are arbitragers buying spot wheat and selling it to the silly fools who want to invest in the higher price futures? The USO oil ETF has suffered from contango, but that hurts investors (ie those carrying oil), it’s good for the producers.

By: csissoko Wed, 24 Jun 2009 22:41:49 +0000 The government report is titled “Excessive speculation in the Wheat market”, which given the growth of long-only funds seems fair to me. It presents findings based on careful research and analysis of the appropriate model to use in evaluating commodities markets (and specifically addresses Nick’s fundamental value “by definition” approach).

You appear to discount the value of qualitative analysis. The Senate, however, is full of lawyers and is unlikely to agree with you in that judgment call.