(Repeats with no changes. The opinions expressed here are those of the author, a columnist for Reuters)
* Chart 1: tmsnrt.rs/2rdqeVV
* Chart 2: tmsnrt.rs/2rMglvl
* Chart 3: tmsnrt.rs/2rMgor1
By John Kemp
LONDON, May 23 (Reuters) - "Backwardation is the solution" to OPEC's problem of how to raise output and revenues without sparking another shale boom, according to the influential oil research team at Goldman Sachs.
Backwardation would allow low-cost oil producers in OPEC to sell their output at a higher price linked to the spot market while curbing growth from shale firms that sell at prices linked to the forward curve.
Goldman's strategy aims to "share growth" between OPEC and shale firms to avoid another repeat of boom and bust in oil prices ("Backwardation is the solution", Goldman Sachs, May 22).
The plan exploits differences in pricing behaviour between low-cost producers in OPEC that do not hedge and higher-cost shale drillers that hedge a substantial portion of their output.
Because they do not hedge, OPEC members' revenues are linked to spot prices but shale firms' earnings and access to capital are more closely linked to futures prices one or two years forward.
Since the end of 2014, Brent and WTI have generally been in contango, with futures prices higher than spot, which has meant shale hedgers have realised higher prices than OPEC non-hedgers.
But if OPEC can shift the market into a sustained backwardation, the situation would be reversed, with shale producers realising lower prices than OPEC.
Goldman estimates that if one and two-year forward WTI prices could be pushed down to $45 per barrel that would be low enough to slow the growth in shale production to a more sustainable rate in 2018.
With a backwardation, OPEC producers would realise spot WTI and Brent prices above this of $50-55 or even $55-60.
Goldman recommends a three-part strategy for OPEC to achieve and then sustain an appropriate backwardation.
First, OPEC needs to extend or deepen its current production cuts to drain excess global inventories from the oil market and shift the whole futures curve from contango to backwardation.
Second, it needs to make explicitly clear that once inventories have returned to the five-year average it will start raising output to keep downward pressure on long-dated futures prices.
If necessary, a verbal commitment can be backed up selling some long-dated futures contracts to keep forward prices under pressure.
Third, OPEC needs to actually increase its production, but gradually and in line with demand, to keep inventories stable and the backwardation in place.
"This seems quite a tall order", Goldman admits, requiring close coordination between OPEC and Russia and within OPEC itself.
"But it is important to note that this was how OPEC proceeded during the 1990s, with steady production growth, decent compliance with output quotas and a persistent backwardation outside recessions".
Goldman is not the only market participant urging OPEC to focus on shifting the futures prices from contango to backwardation, though its strategy has been worked out in more careful detail than most others.
Contango is normally associated with an oversupplied oil market and high and/or rising levels of inventories while backwardation is associated with undersupply and low and/or falling stocks.
Contango makes it profitable for crude traders to store large volumes of crude oil through "cash and carry" strategies which are no longer possible when the market is in backwardation.
Many observers have therefore focused on backwardation as an essential element of any plan for rebalancing the oil market.
Contango is sometimes blamed for causing crude traders to build up stockpiles and causing the oil market to carry excess inventories.
But contango is a symptom of an oversupplied market with abundant inventories, not the cause. Likewise backwardation is a symptom not the cause of an undersupplied market with tight stockpiles.
Holbrook Working of Standard University's Food Research Institute first explained the relationship between stocks and futures prices over 80 years ago ("Price relations between July and September wheat futures", 1933).
Working examined the relationship between stocks and price spreads in the wheat market, and he wrote about positive and negative price spreads rather than contango and backwardation.
Working showed that a large positive spread (contango) was associated with high levels of wheat carried over from the previous crop year, while a negative spread (backwardation) was the result of low stocks.
"Elevator operators who hedge must depend chiefly on spreads between futures for profits from the storage of wheat," Working wrote.
"The elevator operator has little inducement to carry heavy wheat stocks into the new crop year unless September wheat is selling or is expected to sell at a higher price than July wheat."
"The difference between the price at which he buys in his hedge in the July future and the price at which he replaces it by sale of September wheat provides the elevator his chief or only return for storage of wheat."
Stock levels drive price spreads, not the other way around. Spreads do not cause grain elevators to amass stocks but are the consequence of them having to store so much extra wheat.
In the case of oil, market rebalancing will be accompanied by a shift from contango to backwardation, but that will be a symptom of falling inventories not the cause.
OPEC should focus on reducing oil inventories to a more sustainable level and stop worrying about the issue of contango versus backwardation, which will take care of itself.
Goldman is probably correct that contango has favoured shale hedgers at the expense of non-hedging OPEC producers, and a shift towards backwardation will eliminate some of that advantage.
But any shift in the structure of prices (contango/backwardation) is also likely to be accompanied by a change in their level (both spot and forward).
Traders and analysts often distinguish between changes in the structure of prices ("curve") and changes in the outright price level ("flat price").
But while this is analytically convenient, it is a simplification. In reality changes in the shape of the curve and flat prices tend to track one another.
The shape of the curve and the level of prices are not perfectly correlated but they are not wholly independent either (tmsnrt.rs/2rdqeVV).
In the last 20 years, every major shift in oil market structure from contango to backwardation has been accompanied by a rise in flat prices (tmsnrt.rs/2rMglvl).
Backwardation is associated with falling inventories and a tightening market so it is logical that it should also be associated with an increase in flat prices as well (tmsnrt.rs/2rMgor1).
If OPEC succeeds in draining global stocks and pushing the market into backwardation it will be hard to prevent forward prices rising and boosting shale firms too.
Goldman acknowledges that rising long-dated prices as the market shifts into backwardation are one of the risks to its strategy:
"An initial rally in spot prices that would drag along with it long-dated prices, would allow U.S. producers to hedge at a more comfortable oil price and undermine the impact of backwardation at a later date."
It is just about possible to imagine a scenario in which OPEC shifts the market from contango to backwardation, pushes spot prices slightly higher and forward prices slightly lower.
But it would require deft market management and enormous amounts of luck, and the benefits to OPEC's members would be pretty marginal.
OPEC would be better off focusing on a simple strategy of draining excess global stocks, then growing its output in line with forecasts of market demand, leaving flat prices and the curve to take care of themselves.
OPEC's best strategy is simply to maximise its output and prices subject to the need not to ignite a new and unsustainable boom in U.S. shale output.
This is essentially what OPEC members would have done anyway without worrying about complexities such as the degree of contango and backwardation.
Editing by David Evans