November 20, 2011
Implications of the recent rise in oil prices
The price of West Texas Intermediate has risen almost $10 a barrel since the start of September, and briefly bumped back above $100 a barrel this week. Here's why I think that development may not be as worrisome for the U.S. economy as it might sound.
The first point to be clear about is what we mean by the price of oil. Two of the most popular benchmarks are West Texas Intermediate, which is a light sweet crude whose price is quoted for delivery in Cushing, Oklahoma, and Brent, which comes from the North Sea. The crudes are similar in terms of quality, and historically traded for a very similar price. But a year ago the two prices began to diverge, differing by as much as $28 a barrel at the beginning of September.
That divergence resulted from a surge in production from Canada and the northern United States, coupled with a lack of transportation infrastructure necessary to move the oil from where it is now being produced to refiners on the U.S. coasts. For this reason, those refiners have been paying a much higher price for crude oil delivered by ship, with the Brent price representative of what U.S. refiners have been paying to import the oil from overseas.
The Keystone Gulf Coast Expansion Pipeline would today be providing some of that needed transportation capacity if the Obama Administration had given it the ok when the plan was first submitted for approval 3 years ago. However, the White House recently announced that it still needs at least another year to review the plan.
But there are other ways to ship the oil to where the market is signaling that the product is most needed. The Seaway Pipeline has rather astonishingly been transporting oil from the Gulf Coast, where it is expensive, to Cushing, where it is cheap, presumably because ConocoPhillips, part owner of the pipeline, wants to protect its Midwest refining margins. Craig Pirrong looked at this and suggested the logical solution would be for oil producers to buy the pipeline outright from Conoco and invest in the infrastructure adjustments necessary to get the oil flowing through the pipeline in the opposite direction.
And this week Enbridge announced it had entered into an agreement to pay $1.15 billion to buy out Conoco's interest in the Seaway Pipeline, and further announced its intention to make pump station additions and modifications necessary to use the pipeline to transport oil from Cushing to the coast. Enbridge anticipates that it could be ready to move 150,000 barrels per day from Cushing to the coast by 2012:Q2 and 400,000 b/d by 2013.
Although oil won't be flowing for a while, the announcement itself would be expected to cut into the Brent-WTI spread. This is because if someone in Cushing knows there's going to be a better market to sell the WTI to next year, it makes sense to put more into inventory today (driving today's WTI price up). Likewise, if someone on the coast knows there's going to be a cheaper alternative next year, it makes sense to buy less for storage there (driving today's Brent or equivalent price down). If we look at the broader trends since the start of September, what's happened is that the WTI price is up about $9 while the price of Brent is down almost $4.
Because gasoline is transported using different infrastructure from the crude oil, the differences in the cost of crude did not translate into equally dramatic regional differences in the cost of gasoline. Whenever transportation facilities are adequate, we'd expect the law of one price to hold. If you can transport gasoline across U.S. states, the refined product should sell at a similar price as a result of physical arbitrage of the gasoline market. A big effect of the Brent-WTI spread was thus to raise refiners' margins in the central U.S., with the price of gasoline in the U.S. tracking Brent more closely than WTI since the two prices diverged.
It's worth noting that despite the big run-up in WTI, the average retail price of gasoline has followed Brent down, being about 30 cents per gallon cheaper than at the start of September. What matters most for U.S. consumers at the moment is the cost of oil imported by tanker. Insofar as the latest developments put downward pressure on that, the seemingly paradoxical result is that a higher cost of WTI may actually mean lower costs of gasoline for U.S. consumers.
But we've put off the problem long enough, and the size of the challenge is big enough, that more infrastructure is still needed. The Wall Street Journal reports:
U.S. oil production is expected to grow to 6.4 million barrels a day by 2016 from 4.2 million barrels a day today, according to data provided by Bentek Energy, a consultancy. And oil producers in western Canada are expected to ratchet up production to 3.5 million barrels a day by 2015 from 2.8 million in 2010, according to the Canadian Association of Petroleum Producers.
There are conflicting reports in the media (, ) as to whether Enbridge, now that it controls the Seaway, still plans to build the Wrangler Pipeline that the company announced in September, which could transport an additional 800,000 b/d from Cushing to the coast. But even if Wrangler gets canceled, the WSJ also reports:
Magellan Midstream Partners LP still plans to reverse its 700-mile Houston-to-El Paso Longhorn pipeline to deliver up to 225,000 barrels a day of crude oil from west Texas to the Houston by mid-2013, said company spokesman Bruce Heine.
The President may wish to declare that the Atlantic Ocean should be 10 feet lower at New York than it is in Florida. But the water will find a way to get there anyway.
And markets are going to figure out how to sell oil to the highest bidder.
Posted by James Hamilton at November 20, 2011 06:22 AMdigg this | reddit
The WSJ article has a mistake. US crude + condensate production in 2010 was 5.5 mbpd (EIA). If we go back to 2004, prior to the Gulf Coast hurricane damage, the US produced 5.4 mbpd. We will probably average about 5.7 mbpd in 2011.
The US remains dependent on imports for 60% of the crude oil that we process in US refineries. And the annual year to date global price data suggest that we needed a significantly higher annual oil price this year to balance supply & demand. Brent averaged $97 in 2008, but the 2011 average price through October is $112.
In my opinion the global oil price is an indication of an ongoing decline in what I call Available Net Exports (ANE), or the supply of global net exports of oil that is available to importers other than China & India. ANE fell from 40 mbpd in 2005 to 35 mbpd in 2010, an average volumetric decline of one mbpd per year.
I suspect that ANE will be down to between 15 and 21 mbpd in 2020, versus 40 mbpd in 2005.
Posted by: Jeffrey J. Brown at November 20, 2011 07:37 AM
The United States is the most energy dependent economy in the world and yet Hamilton doesn't find rising oil prices worrisome.
Posted by: Ronald at November 20, 2011 08:06 AM
Regarding the surge in Canadian production, BP puts the 2005 to 2010 increase in net oil exports from Canada at 250,000 bpd (total petroleum liquids). In order to fully offset the 5 mbpd 2005 to 2010 decline in Available Net Exports (ANE), we would have needed the incremental increase in net exports from 20 exporting regions like Canada.
Note that of the 12 net exporters that showed increasing net oil exports from 2005 to 2010 (21 of the top 33 net oil exporters showed declining net exports), only 2 were located in the Western Hemisphere, Canada & Colombia. However, combined net exports from Canada, Colombia, Mexico, Venezuela, Argentina and Brazil* fell from 5.1 mbpd in 2005 to 4.0 mbpd in 2010.
*I included Brazil because although the BP data set shows that their net imports of petroleum liquids are increasing, the media seem to generally believe that Brazil is a net petroleum exporter.
Posted by: Jeffrey J. Brown at November 20, 2011 08:27 AM
Loved this article Mr Hamilton.. Please keep these great analyses coming.. the posts are so simple and succinct.. a good resource for regular folks trying to make sense of the world. Thanks :)
Posted by: del at November 20, 2011 11:13 AM
Bentek numbers appear to be onshore oil production (Alaska + lower 48), for July, if we use EIA data. This production is up 540 kbpd over Oct. 2010 levels, consistent with their forecast. Can we put together five years of growth like this? Have to wait and see.
Incidentally, global oil production is up 2 mbpd Oct 2011 over October 2010. And the oil price is still $110 for Brent. Just what is the underlying demand growth for oil?
Posted by: Steven Kopits at November 20, 2011 11:21 AM
"Incidentally, global oil production is up 2 mbpd Oct 2011 over October 2010."
I assume that you are referring to the IEA number for total liquids (inclusive of biofuels) for October, which of course can only be an initial estimate, and which is subject to revision.
The EIA number for total liquids shows flat year over year production for July, and it appears that the average crude + condensate number for 2011, through July, is a little below 2010.
I think that the average annual number is a lot more meaningful, since it tends to average out monthly distortions due to inventory changes.
In any case, as noted above, I think that the primary contributor to 2011 Brent prices averaging $112 through October, versus $97 for 2008, is the ongoing decline in the supply of oil available to importers other than China & India, having fallen at about one mbpd per year from 2005 to 2010.
Posted by: Jeffrey J. Brown at November 20, 2011 12:56 PM
JDH, Steven and Jeffrey,
You guys have good handle on the energy market. Can you share your thoughts on the outlook for world/regional economic growth when world economic activity returns to a normal level? Also, what happens to energy prices and growth if/when the world business cycle enters an above trend expansionary period.
Posted by: tj at November 20, 2011 02:31 PM
if the Obama Administration had given it the ok when the plan was first submitted for approval 3 years ago.
If the plan was submitted three years ago, it should have been approved by the Bush administration. Or could Obama have approved it as President-Elect?
Posted by: athEIst at November 20, 2011 05:58 PM
When reading the IEA report reader may quickly be relieved, no need to be loaded with sophisticated economics theories but trivial business laws will do "charge as much as the market can bear".That is good as it may improve savings on general consumption of oil and oil byproducts and the development of substitute energy sources.Economists may add it could affect general consumption,automobile production and GDP growth but until March 31st the export item heating oil may contribute to improve the US balance of trade.Stable margins for the refiners (IEA http://www.eia.gov/forecasts/steo/pdf/steo_full.pdf) are providing for industrial justice to a line of business that was marginalised in the medieval age of oil production.
The same IEA report does nor fail to bring a message of hope for peace in the middle east.
"However, there may be downward price pressure if Libya is able to ramp up oil production and exports sooner than anticipated.
Oil prices continue to face upwards price pressure because of supply uncertainty resulting from ongoing unrest in the oil producing regions of the Middle East and North Africa and not to forget the gigantic turmoil in Europe that may affect money supply and the course of interest rates.
"Regional turmoil, particularly in Syria and Yemen, exerts additional pressure on the non OPEC outlook and on global oil prices"
All in, LBOs on oil related business should see better days, real negative interest rates improved margins,and higher consumers prices should not be negative factors when it comes to their ability to service their debts.
Since "Nothing is lost, nothing is created, everything is transformed" how will be the reading of this item?
Income Receipts on U.S. Direct Investment Abroad
Posted by: ppcm at November 20, 2011 07:55 PM
Some of the Peak Oilers have differentiated between a problem, which has solutions, and a predicament, which has no solution. One can try to solve a problem, but one has to cope with a predicament.
In my opinion, the fundamental math underlying net (oil) export declines is forcing us to to cope with the predicament of Peak Exports--regardless of how fond we are of our auto centric, suburban way of life. Even within Peak Oil circles, I suspect that only a relatively small number of people really understand the nature of net export declines.
Let me put it this way.
In 2005, Saudi Arabia consumed 18% of their total liquids production (BP). In 2010, they consumed 28% of their total liquids production. If we extrapolate this rate of increase in their ratio of consumption to production, they would consume 100% of their domestic production in 14 years.
At the 2005 to 2010 rate of increase in Chindia's combined net oil imports, as a percentage of Global Net Exports, Chindia would consume 100% of Global Net Exports in 19 years.
And yet, virtually no one is addressing the ongoing, measurable decline in Global Net Exports (GNE) and in Available Net Exports (GNE less Chindia's net imports).
Following are some links to articles.
Very good overview of the global problem:
In the world, at the limits to growth
by David Korowicz
Two articles on "Net Export Math" follow:
Peak oil versus peak exports
Egypt, a classic case of rapid net-export decline and a look at global net exports
My "ELP" advice from four years ago:
The ELP Plan: Economize; localize & produce
Posted by: Jeffrey J. Brown at November 21, 2011 04:39 AM
IN addition to the long term impact on demand at Cushing the transfer of the pipeline, raises the demand at Cushing immediately as the refineries at Bartelsville and Ponca City now have to draw from the WTI pool. One must assume that once the pipeline deal closes the line will shutdown for modifications so the demand change should happen fairly shortly.
On a related matter they just opened a 100,000 b/day rail oil terminal in Western ND, If you look at a rail map that oil is going to the gulf coast because the rails have more capacity that way then to Cushing.
Posted by: Lyle at November 21, 2011 05:58 AM
Another great post with information that is not being adequately reported elsewhere. You are cutting edge.
A portion of the Brent price is the risk factor of the ME supply. Libya is more pacificed and the supply of the area seems more secure than a few months ago. It makes sense that the price of Brent would decline. An attack by Israel on Iran could throw prices into a tizzy.
I do agree that the supply of WTI is greater than the transportation systems to move it and that drove the price down but with the decline of the dollar the normal price of WTI with reasonable reserves would be above $100/bbl. Both the decline in the Brent price and the increase in the WTI price are simply price moves to move the prices toward equilibrium.
Posted by: Ricardo at November 21, 2011 06:46 AM
JDH The Craig Pirrong article that you linked to claimed:
"through February PADD II refiners were working at mid-90s utilization, PADD III in the low 80s, although that differential narrowed in the last couple of weeks with Midcon utilization falling into the mid-80s, probably due to the rise in crude costs resulting from the Mideast turmoil."
He wrote this back in March 2011. But when I look at the EIA's operable utilization rate for PADD II and PADD III, I don't find anything of the sort. It is true that in February 2011 PADD III rates dropped a lot, but this was clearly an outlier and things returned to normal in March. One of the comments on his blog offered an explanation for the February drop-off.
Over the last couple of years the two regions have had almost identical operable utilization rates. And for Aug 2011 (the most current month in EIA's database), the PADD III rate was 94.1% compared to the PADD II rate of 90.7%. So I guess I don't understand Pirrong's point. My little brain tells me that a 94.1% utilization rate refining capacity in the Gulf ought to be more of a concern than the amount of crude getting to the Gulf. Based on all of the arguments for the Keystone Pipeline I would have expected to see very high PADD II utilization rates and declining rates in the Gulf region. But that's not what the EIA data shows at all. If anything the Gulf utilization rates are very high and would seem to suggest an argument for increased refinery capacity in the Gulf.
I'm also not quite convinced of Pirrong's Coase Theorem argument. He is calling the cost of reversing the Seaway Pipeline a "social" cost because of the loss of consumer and producer surpluses. Well, okay; but let's remember here that the agents are refineries and pipeline owners and not consumers in the normal sense of the term. As you noted in your post, the effect on the final consumer is likely to be lower gasoline prices. There may well be a deadweight loss that does not accrue as a benefit to either the pipeline owners or the refineries, but that does not mean there is a deadweight loss to society in general...especially when you consider externalities.
Posted by: 2slugbaits at November 21, 2011 01:06 PM
Bakken, Eagle Ford, Wattenberg, Utica
House prices always go up.
US LNG imports will grow dramatically.
US oil production is only going one way.
These were all conventional wisdom until recently - the last one still is.
Beware of static analysis. The market responds to price signals if you let it, both supply and demand. Technology changes - and people find new ways to do things - ask the post office about that.
Posted by: Equityval at November 23, 2011 01:26 PM