From one of the text books (“Investments” by Bodie, Kane and Marcus : McGraw-Hill) I used in one of the finance classes I took.
“”Future Prices versus Expected Spot Prices
So far we have considered the relationship between futures prices and the current spot price. One of the oldest controversies in the theory of futures pricing concerns the relationship between futures price and the expected value of the spot price of the commodity at some future date.”
The authors discuss three traditional theories which have been advanced; the expectations hypothesis, normal backwardation and contango. They conclude this section of the text with a discussion of modern portfolio theory which they say is understood to subsume the three theories mentioned above.
The book’s glossary states that “Normal backwardation theory [h]olds that the futures price will be bid down to a level below the expected spot price” and that “Contango Theory [h]olds that the futures price must exceed the expected future spot price”
[Their explication of this topic covers three pages so because of (i) copyright concerns and (ii) my time constraints I am going to paraphrase to give the gist of their discussion,]
The expectations hypothesis holds that Fo = E(Pt). It is simplistic and unrealistic and is likely to be accurate only transiently, if at all, apart from at the expiry date of the futures contract.
With respect to the other two theories, the authors ask the reader to consider grain growers and grain millers. Growers desire to sell their grain for as much as they can, and are seen as “natural” short hedgers, who will agree to provide grain in the future at an agreed price as per the futures contract specifications. Speculators who hope to profit from taking the other (long) side will only do so if the futures price has been bid down so that Fo < E(Pt).. This situation is termed normal backwardation. It is depicted in the text as an upward sloping line wherein the futures price is below the expected (i.e. in the future) spot price and rises until the two prices match at the maturity date.
In direct opposition to this theory, contango holds that “the natural hedgers are the purchasers of a commodity, rather than the suppliers”. Purchasers will hedge long. They hope to pay as little as possible for the grain they need to buy, and in this case speculators will only take the short side of the contract if the futures price has been bid up so that Fo > E(Pt) to offset their risk. This (contango) is depicted in the text as a downward sloping line wherein the futures price is above the expected (i.e. in the future) spot price and falls until the two prices match at the maturity date.
Any commodity futures market will have both producers (short hedgers) and purchasers (long hedgers), plus the presence of speculators (liquidity providers). Whether the market is in normal backwardation or in contango at any given moment will depend upon the relative numbers of these factions and which side (short or long) is dominating. “The strong side must pay a premium to induce speculators to enter into enough contracts to balance the “natural” supply of long and short hedgers”
So regarding the recent oil futures volatility and price collapse, the long and short of it is that holders of contracts which obliged them to take physical delivery of oil were panicking and were willing to pay anybody to take the contracts off their hands.