Its hard to take on a model when you do not know how it works, although I would love the challenge based on the obvious events in Oil and other commodity markets showing speculation at work.
However my understanding (please correct me if wrong) of this model is as follows:
1. Only the nearest contract month in considered 2. Transactions in commercial interests are not considered
Dealing with point 1 there, speculators are not interested in dealing in the contract closest to delivery, especially in oil where all contracts are very liquid. Action in the other contracts influences the spot price and thus escape the model. This is especially true when you consider a lot of smaller investors gain access to the oil market through indexes representing commodity prices who invest in the actual product - they simply can not deal in the contract closest to delivery.
2. Simply because an entity is classed as a commercial interest does not mean they are not trading for speculative purposes. If you have oil prices moving up and a company finds the need to hedge against future oil price rises because of the ever increasing prices then I would class this as speculation.... because when prices fall like they are now they are less likely to hedge and a whole lot of demand is withdrawn from the market causing the price to crash.
In addition their own statistics show the percentage of trades in the contract nearest delivery (I believe) doubling from 18% to 35%ish the last several years.
The difference in supply during the 1929 crash was only 3%. An increase in speculation of this amount, persistent long term is enough to soak up the supply in any market.
In my opinion the people who made the conclusions of this report really don't have much intricate understanding of the dynamics at play.
