Contango is not necesarily considered the "normal" state of affairs for all commodities.  Grains are harvested once per year, and in a normal year that supply must be allocated over the coming year.  Higher prices for the contracts late in the crop year are the market's way of encouraging storage.  In a short crop year, price are higher in spot markets and nearby months to discourage storage.

Crude oil, however, is produced continuously.  It is not at all unusual for the crude oil futures market to be in either contango or backwardation.  Both conditions have prevailed for extended periods over the last two decades.  The point still stands, however, that a backwardation reflects a tight current balance between supply and demand that is expected to ease going forward, and a contango would relfect  the opposite situation.

The thing that is most interesting is that the distant contract prices can be interpreted as the market's long run forecast regarding the supply and demand balance.  As such, distant contract prices are much less volatile than spot prices and the prices of nearby contracts, which reflect every short-run production/distribution interruption, the current year's weather, etc.  The market anticipates such short-run issues working themselves out in some limited amount of time.  Distant contracts' prices, by contrast, reflect more signal and less noise.

The prices for delivery of crude oil five years in the future were very stable at around $20 per barrel for most of the history of the NYMEX contract prior to 2003 (I wish I knew how to post a spreadsheet chart that I created that shows this).  Since that time, the five-year forward price has gone up fairly steadily to its current level of around $65 per barrel.  The market concensus regarding the long-run balance betweeen supply and demand is changing substantially.  There is the big story.

I'd love to see that chart.  Perhaps one of the TOD staff could post it for you?  If not, you could send me the spreadsheet and I'll post it.

I agree - it's the rise in distant future price that's a real sea change.  No more "oil will be back down to $20/barrel" any more.  

I don't think that the editors want to do such a thing for everybody.  I looked at your user profile, but did not find an email address.  Email me at subtr4ct-at-gmail-dot-com if you could post the chart and could reply with the Excel file that has the chart.
Okay - I think I figured out how to post the chart...

The prices are for NYMEX crude oil futures contracts delivering about 3 years in the future in the early 1990's, as the futures did not trade 5 years out back then.

subtr4ct

Thanks!  Nice chart.  

If the end is chopped off, here's the link to view it directly:

http://i78.photobucket.com/albums/j100/subtr4ct/oil.jpg

It really spikes up, starting around the beginning of 2005.

That's a great chart, thanks for posting it. It's amazing how flat it is. One question, are the dates shown as the contract delivery dates, or as the trading dates? So for example the firet one, 11/23/1990, is that the price in 1987 for a contract delivering in 1990? Or is it the price in 1990 for a contract delivering in 1993?

It would be great to compare it to then-current oil prices to see how it compares, but I need to know what the dates mean.

Halfin, The dates are the time of trading in the future 5 years forward.
Glad you find the chart interesting.  I first made a chart like this in July of last year, and it led me to really put my money where my mouth is.

The first observation is the settlement price on 11/23/1990 of December 1993 futures.  The last point is yesterday's closing price of Dec. 2011 futures.

Shit! OK, another coffee-spitting moment!

This graph is one of the most compelling illustration of the "Peak Now" that I have seen this year.

This really helps to debunk the view that the current 'plateau' is just another little hiccup in the march on up to 120Mbd in 2030.

Thanks a lot for posting this graph.  I hope Stuart can add this to his "why we're probably peaking now" page.

Sorry.  Go ahead and swollow your current sip...  Okay, the picture for natural gas is very similar:

Note that I have used a dollars per 100,000 BTU price (not the conventional dollars per 1,000,000 BTU price) to make the two series of comparable magnitude.

I don't suppose anyone here thinks the price of oil in 5 years will be anything near $65, which sorta suggests a buying opportunity, or am I reading this wrong?
good buy if you have deep enough pockets and youre patient. long term, i wonder what the chances are of NYMEX ceasing to operate or certain classes of traders being restricted. anyone know what might cause either of these things to happen, besides the obvious 'end of the world' stuff?
I have been wondering this too.  If the market was left to its own devices, I think today's prices for crude five years out are a steal.

But governments could intervene and your options may as well be worthless.  But peak oil in combination with inflationary fighting measures could so devastate the economy (via deflation) that oil five years could cost only half as much as it does today.  But one shouldnt rejoice if average income has fallen to one tenth of today's level.

Has anyone else thought about this?  Peak Oil decimating the economy to the point where the price of oil falls simply because of deflationary pressures (think cascading unemployment reducing demand for virtually everything which in turn reducing pressure on scarce resources). Add in potentially restrictive government restrictions like gold confiscation or protectionist terrifs or high interest rates or permitting employers to unilaterally cut salaries).

Or is the inflationary route pretty much guarenteed?

Deflation is my primary concern, as I have attempted to explain here many times over the last several months. I would expect oil prices to fall (for a while) as a result of severe economic contraction, but I agree with you that purchasing power would be falling even more quickly. Oil would nominally be cheaper, but less affordable than it is now. I don't see deflation as being confined to the US, so a similar dynamic could be widespread.
But governments could intervene and your options may as well be worthless.

What government intervention has been or could be used?

I was just imagining ordering the halt in the trading of oil futures.  Hasn't happened, but it doesnt mean it couldnt.
Can you tell me where (link?) you got the historical prices of the futures contract?  Thanks!
All data in both charts that I posted are from Primark-Datastream, a subscription data service.  I expect that it is fairly expensive, but I don't actually know.  Such data are available from numerous services, however.  I don't of any free sources for historical futures data.
I have the futures data - if you email me I can send it to you
I think that the signal-to-noise concept is very interesting.  Most striking to me is that there is such a jump in near-term futures prices.  At the moment, there is a jump of about $3/barrel from the spot price to the December '06 price, and then a gradual--if uneven--decline each year one goes out beyond '06.  This initial jump makes sense to me, and I think it is the most accurate forecast of the "terror/weather" premium, because a storm or attack is assumed to be a relatively short term interruption.  So for that reason the prices for December delivery are higher, as they incorporate all this noise--political maneuvering, terrorist risk, weather, etc., but the longer-term prices represent the market-consensus on signal--that is, what the future supply/demand fundamentals will look like.

To me, if (or once) the market accepts Peak Oil, then the degree of contango--especially in the 3-5 year out range--will represent the best market information on interaction of declining production and demand destruction.  This will, I'll argue, not actually be "contango" per se, as it will not be caused by the same market forces that cause "standard" contango--it might look more like contango in S&P500 futures back in 1995.  That is to say, it will move upward sharply based on future assumptions of declining production.  If global decline proceeds at 5% beginning now, how much could we expect to see the futures price increase year-on-year?  10%?  I have no idea how to begin calculating this.  It does strike me, though, that contango is constrained at some point by arbitrage:  if contango is at 10% a year moving forwards, that will exert a strong pressure on current spot prices because of increasing demand from arbitrageurs who sell a future, buy and store the oil, and deliver.  Depending on how well inflation risk can be hedged with other instruments, and the performance expectations of other markets, I suspect that contango exceed 15% a year because of arbitrage pressure...  any historical data or thoughts to back up this concept of contango and arbitrage dragging spot prices up by the bootstraps?

I'm sure you're right, Jeff, that the market structure won't allow oil futures to increase too much year over year. If oil producers see that next year's prices are substantially above this year's, it is easy for them to back off on current production in order to have more for next year. That will raise current prices and potentially lower future prices, bringing them back into balance. There is arguably evidence for this behavior today, in terms of OPEC's failure to aggressively invest to increase present-day oil production.

It's harder to say exactly how big the increase could become before this is a major factor. And of course we might see short term fluctuations that are substantially larger. In principle any rate of increase above the prevailing interest rate could trigger the effect (this is the Hotelling model for production of exhaustible resources). That would be 5-6% per year. In practice it could probably be a little more than that but I guess I'd be surprised if it stayed above 10% per year.

Note that this effect looks to the naive eye like greedy speculators are artificially driving prices higher than they should be under conditions of supply and demand. And note too that we have heard exactly this charge in recent months, as oil prices have stayed high despite a short-term glut as evidenced in growing storage inventories. However Hotelling showed that this production profile is actually optimal in terms of maximizing the net economic value of the oil.

We should be glad when this happens; by slowing down current oil consumption we leave more for the future. Some people argue that this is not enough, we should tax it to make it even more expensive, but this is economically inefficient and will cause excess and premature consumption declines. It makes people poorer today and gives them fewer resources with which to deal with future challenges. The best method is to let the market predict future prices, let those future prices drag up today's prices, and trigger conservation on that basis.

The effect should work in the other direction, too, BTW; if future oil prices are not rising at a decent pace (i.e. keeping pace with inflation plus a few percent) then we would expect to see higher production today, leaving less for the future. This effect may have been operating in the past but we will likely see a transition away from it as we approach the peak.

The problem with the argument that speculators will limit the degree of contango by selling the longer term contract and buying the near contract, taking delivery  and storing the oil against their long term short is that few speculators - if any - have the operating ability to take possession of large quantities of oil and keep it stored for years.   Maybe that will develop, but I kinda doubt it.