(Repeats Dec. 18 item. John Kemp is a Reuters market analyst. The views expressed are his own)
By John Kemp
LONDON, Dec 18 (Reuters) - "If we consider the actual basis of this (intelligence), how unreliable and transient it is, we soon realise that war is a flimsy structure that can easily collapse and bury us in its ruins."
German military theorist Carl von Clausewitz was writing about the role of intelligence in warfare, but his insight about unreliable information is equally applicable to traders, investors and business leaders trying to understand whether oil prices have fallen far enough or have further to drop.
"Many intelligence reports in war are contradictory; even more are false, and most are uncertain," Clausewitz explained in his 19th-century treatise "On War". "The effect of fear is to multiply lies and inaccuracies. As a rule, most men would rather believe bad news than good, and rather tend to exaggerate the bad news."
Only experience and a sense of the laws of probability can help an officer make sensible decisions amid a welter of conflicting and often inaccurate information.
Modern military forces, led by the United States, expend enormous amounts of money on the acquisition and processing of information about the state of the battlefield in real time to improve decision-making.
But for the most part, decision-makers in commodity markets must operate in an information vacuum where data is mostly estimated, incomplete and out of date.
Not only is the future unpredictable, we are not even sure what is occurring at present or has happened in the recent past.
Common sense suggests the near-halving of oil prices since June should already be affecting decisions about future supply and demand - everything from drilling plans to the type of cars consumers buy and their levels of discretionary driving.
But current information on almost every element of the supply-demand picture is seriously out of date - assuming it is accurate. Even in the United States, which produces the most comprehensive, accurate and timely statistics, the bulk of such data is up to three months old.
Data on drilling permits and drilling activity is available on a daily or weekly basis from state regulators and oilfield service companies such as Baker Hughes. But reasonably accurate and comprehensive information about oil production and the consumption of refined fuels is mostly more than two months old.
The most recent comprehensive estimates for oil production and products supplied to domestic customers published by the U.S. Energy Information Administration relate to September, when oil was above $90 a barrel, compared to some $60 in recent days.
Data on vehicle sales is available a little faster, for October, but the most recent estimates for driving relate to September.
Information on supply and demand in other major oil-producing and consuming countries is available only with much longer delays and far less reliable.
Delays are a problem at any time, but they become critical when the market is approaching a turning point and prices are changing quickly.
Delays contribute to major analytical errors. In 2007/8, the majority of oil analysts failed to spot signs of demand destruction even as crude doubled from $75 to almost $150.
Because demand destruction was not yet evident in the data, many analysts reasoned (falsely) it was not happening and prices would have to rise even further to restrain consumption.
As prices jumped to $100, then $125 and even $147, many analysts were still predicting they would need to go even higher to bring the market back into balance.
In fact, prices plunged. In retrospect, with better data available, it became clear that consumption had been falling briskly in the advanced economies for up to a year.
Some analysts and investors risk making the same mistake in 2014. Because there is so far no sign of a slowdown in oil drilling and production in the time series data, or a pickup in fuel consumption, it is tempting to argue that prices have not yet fallen far enough and need to drop even further.
But that would be wrong on several counts. First, the effect of recent price declines is still filtering through to producers and consumers: the full impact might not be felt for another six to 12 months. Second, the data series are significantly out of date, providing a good picture of the market in September, before most of the decline occurred.
In this information vacuum, it is necessary to employ Clausewitz's advice about judgement and probability to make sense of the market and prices.
The magnitude of the price change over the last six months - which is comparable to 1985/86, 1997/98 and 2000/01 - is probably sufficient to force expected supply and demand back towards balance by the latter part of 2015.
Prices are now in the lower half of the range cited by most experts as the full breakeven cost of U.S. shale production. Producers in the United States and around the world have already started to scale back investment in 2015, cutting or cancelling drilling programmes.
In the last month, there have been announcements about reduced investment from Bakken driller Continental Resources , Canada's Cenovus, ConocoPhillips and Britain's North Sea producers.
Britain's North Sea oil industry is "close to collapse", one executive told the BBC in an article published on its website on Thursday ("North Sea oil industry close to collapse").
"In terms of new investments, there will be none, everyone is retreating, people are being laid off at most companies this week and in the coming weeks. Budgets for 2015 are being cut by everyone," Robin Allan, chairman of the independent explorers' association Brindex, was quoted by the BBC as saying.
More producers around the world will follow in the next few weeks and in January, when fourth-quarter results are released. It will take time for all these cancellations to affect supply, and even longer to be visible in the data, but the adjustment is under way.
On the demand side, there are signs U.S. consumers are starting to buy larger, less fuel-efficient, vehicles. If fuel prices remain relatively low, the amount of discretionary driving will increase.
No one can predict with certainty how much lower oil prices might go in the short term, as the market gropes for a floor.
But with prices down almost 50 percent, shifts in supply and demand already happening, and most data more than two months out of date, experience and probability counsel that adjustment is on the way even if not yet visible. (Editing by Dale Hudson)