August 07, 2006
Bad news from Alaska
The response of oil prices to the news from Alaska has been more modest than I was expecting, with the September NYMEX crude contract currently up about $2 to $77/barrel.
Oil giant BP today released the following statement:
BP Exploration Alaska, Inc. has begun an orderly and phased shutdown of the Prudhoe Bay oil field following the discovery of unexpectedly severe corrosion and a small spill from a Prudhoe Bay oil transit line. Shutting down the field will take days to complete. Over time, these actions will reduce Alaska North Slope oil production by an estimated 400,000 barrels per day.
The decision follows the receipt on Friday, August 4 of data from a smart pig run completed in late July. Analysis of the data revealed 16 anomalies in 12 locations in an oil transit line on the eastern side of the oil field.
In response to the inspection data, BP conducted follow up inspections of anomalies where corrosion-related wall thinning appeared to exceed BP criteria for continued operation. It was during these follow up inspections that BP personnel discovered a leak and small spill estimated at 4 to 5 barrels.
Prudhoe Bay is America's biggest oil field, and this account does not sound to me like a problem that's going to be fixed quickly or easily. A 400,000 barrel/day shortfall would represent about 5% of total U.S. crude production (including lease condensates, natural gas plant liquids, and refinery process gains) and 0.5% of the world total. The latter is the relevant number for the effect on the price of oil, though the former may imply a disproportionately bigger impact on the U.S. west coast and balance of payments.
As of the moment, the NYMEX September contract has moved (76.85 - 74.76)/74.76 = 2.8% in response to news of a global production shortfall of 0.4/84.5 = 0.47%, which would be consistent with a short-run demand elasticity of 0.47/2.8 = 0.17. One could certainly defend a lower short-run elasticity than that, implying a bigger price increase. Calculated Risk is predicting a $5/barrel price increase to $80/barrel.
Will the lost Alaskan crude be made up elsewhere? I wouldn't look for it to come from Saudi Arabia, where the most recent data continue to confirm further drops in production. A survey by Bloomberg of oil producers anticipates another 70,000 barrel/day cut in Saudi production to 9.15 mbd for July, while Petrologistics said it might be as low as 9 mbd. Such numbers imply that the Saudis are currently producing 400,000 to 500,000 fewer barrels/day than what they did last year.
The consequences of the Saudi cuts had presumably already been factored into oil prices, and the possibility of a disruption such as that from Alaska may be argued to have been priced in as well in terms of the pre-existing speculative premium. Still, this is not good news at all for oil markets and the U.S. economy.
Posted by James Hamilton at August 7, 2006 11:56 AMdigg this | reddit
Listed below are links to weblogs that reference Bad news from Alaska:
» BP's Prudhoe Bay Pipeline Shut Indefinitely from Knowledge Problem
Lynne Kiesling This is a humdinger: BP's Prudhoe Bay pipeline is possibly corroded and leaking, and will be shut down. Once the field is shut down, BP said oil production will be reduced by 400,000 barrels a day. That's close... [Read More]
Tracked on August 7, 2006 03:13 PM
» It’s all about the oil from The Glittering Eye
The news on oil production is not particularly good. Here’s a cheery item: ANCHORAGE, Alaska — In a sudden blow to the nation’s oil supply, half the production on Alaska’s North Slope was being shut down Sunday after BP Explora... [Read More]
Tracked on August 7, 2006 05:00 PM
The relatively modest price response appears to me to be evidence for the argument that the primary reasons for high oil prices are long-term major concerns, not short-term minor ones. The nice thing about a broken pipeline is that the oil will still be there when the pipeline is fixed, and in a peaceful country, that's just a matter of time.
Posted by: Stuart Staniford at August 7, 2006 12:48 PM
Don't they have to wait until the winter season and frozen ground before they can do construction work? Could this interruption last longer than just a few weeks?
Posted by: Joseph at August 7, 2006 04:40 PM
winterin alaska= oct
Posted by: Anonymous at August 7, 2006 08:39 PM
How long did it take the US to get the USS Yorktown repaired for the Battle of Midway? I believe it was three days.
Hey BP, hire the workers and engineers to fix the existing leak and survey the rest of the infrastructure.
This oil pipeline problem is just the tip of the iceberg when it comes to infrastructure problems in this country (electrical, fuel, roads, bridges, sewers, etc).
Posted by: Grzegorz at August 8, 2006 03:07 AM
If high oil prices are due to "long term concerns" then the futures market should be much higher than todays price, but Dec 2010 crude price is only $73bbl, less than today. I suspect the market reaction is muted as the market probably figures that the problem will be resolved fairly quickly and can be coped with by drawing down the historically very high stocks currently being held.
Posted by: ChrisA at August 8, 2006 03:14 AM
I second ChrisA's view, with a few expansions. There are high inventories of crude and product on the West Coast. There is also (according to private sector and OPEC estiamtes) a roughly $15/bbl risk premium build into crude prices. If an unanticipated problem arises when the risk premium is 5% of the total price, I'd expect a big price response. When the risk premium is 20% of the total price, I'd expect a smaller response. Even though you didn't anticipate a particular problem, you have anticipated problems in general. In some sense, this problem was already anticipated in oil prices, even if we didn't know the exact nature of the problem.
Posted by: kharris at August 8, 2006 06:23 AM
So.... what happens if other not anticipated problems happen? For example, a war against Iran...
Posted by: Joćo Carlos at August 8, 2006 07:09 AM
Look, failure of rational expectations:
Where is price gouging when you need it?
Posted by: Angela at August 8, 2006 07:12 AM
I can claim no special expertise on oil pipeline repair but I have worked on pipe in nuclear plants.
If there were only 16 indications on a complete survey of their piping, I'd doubt that repairs and restart would take more than a month. Patches could fix the 16 min wall finds fairly quickly. A longer term fix would replace the line.
My guess is a month, max and maybe 10 days minimum.
Posted by: Joseph Somsel at August 8, 2006 11:56 AM
Joseph: Latest from BP is that they are going to replace 16 miles of pipeline. If that's the case, then that makes me wonder whether the 16 miles of pipe even exists? Anyone seen 16 miles of pipe just sitting around collecting dust lately? I would presume, in this day and age, that material such as this is produced according to a plan and with a request delivered long before delivery date.
If so, me thinks some estimate of closure through end of year and into 2007 make sense.
That said, I have no experience with pipes other than some random plumbing around the house and some adventures at concerts in the late 70s that I'll not discuss further.
Posted by: T.R. Elliott at August 8, 2006 01:27 PM
If they intend to REPLACE 16 miles of pipe, then it will indeed take a while to get the raw pipe made, inspected, shipped, and installed. If they intend to patch the 16 indications of wall thickness below minimum standards, than that can be done relatively quickly. The bends in the pipes are typical problem areas.
Why they chose to shutdown production on the line while waiting to replace the pipe rather than repair and return to service with eventual replacement is still a mystery. They should have a good estimate of future thinning rates and with a complete wall thickness survey could reliably predict any needed repairs beyond the 16 indications required until the whole line is replaced.
The operators and regulators are very sensitive to leaks (see report of 4 or 5 barrels!) but this sounds excessively cautious.
Posted by: Joseph Somsel at August 8, 2006 04:55 PM
I think you guys that understand nothing about oil production need read what the guys that work for oil companies are saying about the pipeline repair at the Oil Drum site.
Sulfur bacteria can make a lot of damage to metal...
Nothing personal, but it is not good economists think that they know everything about chemistry, engineering and biology. It is better the shoemakers deal only with shoes...
Posted by: Joćo Carlos at August 9, 2006 07:27 AM
Carlos: Huh? Where did that come from. I'm a physicist and engineer by training. And economists can know a lot about oil production. The oil drum is a great place, but you've got a lot of people writing their with no background in oil production. A computational physicist who has taken an interest in it, a sociologist--if I'm not mistaken, or political scientist, etc.
Sorry, but your point makes absolutely no sense at all.
Posted by: T.R. Elliott at August 9, 2006 09:16 AM
I will admit that Carlos has a point - I'm only speculating based on the facts publicly reported here and on my own experience working on pipes for nuclear power plants.
The facts I'm working with could be inconclusive or just plain wrong. But I'm sure that DoE, Interior, SPR, state of Alaska, etc are all over this and have boots on the ground to verify the repair or replacement plans.
If operational regulations are so tight that they require a shutdown until replaced given current conditions, then perhaps the rules will be suspended for a while.
Posted by: Joseph Somsel at August 10, 2006 12:51 PM
I would like to note that in the recent investigations of the North Slope BP oil pipe line leaks that employees have spoken out on the leaks years before they happened. A friend of mine who is now retired from BP was an engineer and made a report showing the needs of corrosion maintenance for the pipes so that they would not spring leaks. What came out of it was in a sense, say it again and you will be fired. So, the engineers turned and looked the other way on the needs of corrosion reports and maintenance. The bottom line is that if there is no whistle blower protection for the employees who know what is going on and could report what is needed to prevent leaks of the pipe line, then they will not report them and the pipe will have more leaks at a later time. No one is going to report what is needed and be fired later for doing so. So the choice is to provide whistle blower protection or have more leaks in the future
Posted by: John Suter at November 3, 2006 03:40 PM