skip to main content

→ Top Stories:
Safe Chemicals
Defending the Clean Air Act

Anthony Swift’s Blog

New Report: Keystone XL tar sands pipeline will increase U.S. gas prices

Anthony Swift

Posted May 22, 2012 in Moving Beyond Oil, Solving Global Warming, U.S. Law and Policy

, , , , ,
Share | | |

One of the most misunderstood issues surrounding the proposed Keystone XL tar sands pipeline is its impact on U.S. gasoline prices. NRDC, Oil Change International and ForestEthics Advocacy released a report, Keystone XL: A Tar Sands PIpeline to Increase Oil Prices, today that take a close look at this complicated issue and evaluates Keystone XL’s impact on U.S. gasoline prices and supply. The study finds that Keystone XL is likely to both reduce the amount of gasoline produced in U.S. refineries for domestic markets and increase the cost of producing it, leading to even higher prices at the pump. Keystone XL’s supporters in the United States cite high gasoline prices as a reason to overlook the project’s tremendous environmental impacts and build the project. There are plenty of compelling reasons not to build the Keystone XL tar sands pipeline – it will expand a destructive extraction process, put our rivers, aquifers and lands at risk of tar sands oil spills, and would increase our dependence on tar sands – worsening climate change and undermining efforts to move to clean energy. In addition to this litany of problems, rather than decreasing U.S. oil and gasoline prices, the Keystone XL tar sands pipeline will lead to even more pain at the pump for American consumers.

So how does a pipeline increase gasoline prices?

First, Keystone XL is going divert crude oil from the Midwest, a region of the United States where refineries are designed to produce as much gasoline as possible from a barrel of oil, to the Gulf Coast of Texas, where refineries have reconfigured themselves to produce as much diesel as possible from a barrel of crude.

These differences are due to recent changes in the world market for oil and refined products. Historically gasoline commanded higher prices than diesel, driven largely by increasing U.S. consumption. That has changed over the last ten years as U.S. gasoline consumption has plateaued, due in large part to increasing automobile efficiency standards, and global demand for diesel continues to rise. Today, diesel commands a higher price than gasoline, particularly in the international market.  

Gulf Coast refineries, which have access to lucrative international diesel markets, have taken advantage of these international trends by reconfiguring their operations to maximize diesel production. Meanwhile, Midwestern refineries, which do not have access to these international markets, have maintained their focus on the U.S. market. To give an example, refineries in the northern Midwest produce almost 22 gallons of gasoline from a barrel of oil, which refineries in the Texas Gulf only produce about 17 gallons. Moreover, refineries in the Texas Gulf are working with better quality crudes, crude which should produce more gasoline than the heavier, more sulfuric crudes processed in northern Midwestern refineries.

Right now, Canada doesn’t have ‘extra oil’ to put on Keystone XL. In fact there are nearly 2 million barrels a day of empty pipeline capacity available to export additional Canadian crude oil – Canada doesn’t produce enough oil to fill it. That means that in the short to medium term, Keystone XL will allow producers to send oil to the Gulf Coast instead of the Midwest. Simply sending 830,000 barrels of crude oil to Gulf Coast refineries instead of to Midwestern refineries will actually reduce the amount of gasoline produced by U.S. refineries by 80,000 barrels a day – that means about reducing the amount of gasoline available to U.S. consumers by about 1.2 billion gallons a year.

 Second, Keystone XL will increase the cost of producing gasoline in the Midwest. Which TransCanada originally pitched Keystone XL to Canadian regulators in 2009, the company asserted that its pipeline would increase the price of Canadian tar sands. TransCanada argued that Canadian tar sands had historically sold on equal terms with Mexican heavy crude, but that increasing supplies in the Midwest had caused Canadian crude to sell at a $3 discount – a discount that Keystone XL would eliminate, increasing revenue to Canadian crude producers.  In 2012, the discount for Canadian tar sands has increased to between $20 and $40 a barrel. Based on TransCanada’s earlier analysis, by eliminating that discount, Keystone XL would increase the cost of crude by up to $27 billion a year.

Thumbnail image for Graph Oil Prices.JPG

This wouldn’t matter for consumers if, as Keystone XL proponents often argue, Midwestern and Rocky Mountain refineries were just using the lower cost for crude to increase their profits. If that were the case, increasing the cost of oil by $20 to $40 a barrel would have little effect on consumers. The problem is, Midwestern refineries aren’t making significant profits. Even with after paying significantly discounted prices for crude oil, Midwestern refinery margins, or revenue per barrel refined, are less than Gulf Coast refineries, which pay significantly higher international prices for oil.

This is because Midwestern and Rocky mountain refineries don’t currently have access to more lucrative international diesel markets. If these refineries are forced to pay the higher international price of oil, they will be forced to cope the same way East Coast refineries have – by reducing their production, further decreasing U.S. gasoline supplies and increasing prices. 

There are many reasons to argue that Keystone XL isn’t in the national interest. A closer examination of Keystone XL’s impact on U.S. oil and gasoline supplies provides one less reason to support the project.

Go here to read the full report.

Share | | |


Mahima SMay 22 2012 06:11 PM

Hi Anthony-

I'm avidly against the building of the keystone XL pipeline, and have had several debates with a friend of mine who thinks it is a necessity. When I shared your article with him, this was his response (copy pasted from an email exchange).
Since you are close to this topic, could you please take a look at his facts to see if they are correct at all?

"Haha he’s intelligent but slightly misleading I think:
1. It may lead to an increase in prices in the US Midwest, but

a. Will definitely lead to a price decrease in global gasoline prices because these refiners in the Texas gulf export a LOT of gasoline..and the emergin markts are where gas prices will roof without this thing.

b. Will probably lead to a price decrease in the Northeast, because there are gas pipes from Texas to New York, but not from IL to NY. And NY/PA/NJ consume about 60% of the US’s gasoline

2. Midwestern refineries are currently MINTING money..i don’t understand what he means when he says they’re not. so much so, that in fact, there is one refinery that BP owns in Indiana that FULLY pays for the cleanup of the BP spill in the gulf. It’s an atrocious amount of money ($5-$7bn) coming from that one refinery every year. Mindboggling.

3. Refineries in the Midwest definitely have access to better quality, cheaper crude oil than those in the TX gulf. I don’t know how that strengthens his point, but he happens to be wrong..

Believe me when I say that I’m super pro-environment and even I think this pipeline is on balance a good thing for the world. but that’s because I think starvation (bit too extreme a word) and economic strife should be valued higher than saving the environment, which I know you don’t necessarily agree with."

BSMay 22 2012 08:20 PM

Fascinating. So a report written by NRDC happens to support NRDC's position on Keystone. Allow me to help you complete the story:

You claim that tar sands production is 1.6 million barrels/day. However, you conveniently forgot to mention that the US imports about 2.3 million barrels/day from Canada, and that amount is growing. (

You also ignore the fact that some of these pipelines may also carry US oil in addition to what they import from Canada.

You also ignore the fact that Keystone XL has been designed to carry both Canadian oil as well as very light Bakken (Montana, North Dakota) oil.

You ignore the fact that we also import Canadian oil via rail because the existing import pipelines don't always get the oil to the places it needs to go. The fact that a pipeline is not operating at capacity doesn't mean anything unless you know WHY it's not operating at capacity.

And as always, your report has a healthy dose of trying to have it both ways. First you claim that we can't import much more oil than we already are, and then you also point out that the price of Canadian oil is only about $70/bbl, which is caused by an excess of supply due to lack of demand (in this case due to transportation bottlenecks). So which is it? Is there a supply glut or are we already importing as much as they can possibly produce?

BSMay 23 2012 08:38 AM

Your readers might also be fascinated to learn that while Canada actually produced about 3 million barrels/day in 2011 and is expecting to produce about 3.3 million barrels/day in 2012.

Much of this production (but not the majority) is conventional light crude which we also import from Canada. I think you forgot to even acknowledge its existence.

Anthony SwiftMay 23 2012 11:09 AM

Hi Mahima,

It's certainly true that there are times one has to make tough choices - but even in this debate, proponents of Keystone XL have not argued that their pipeline will reduce starvation or economic strife. However, if you look at agricultural projections moving forward, there have been compelling arguments that these will be the results of a warming planet.

However, to your questions:

1. This is a complicated issue to unpack - but here's my quick try:

a. First off, outside North America, the world uses primarily diesel, not gasoline – and that’s where much of the increasing demand is coming from. Another point to understand is that Keystone XL will be reducing oil supplies available to the Midwest and Rocky Mountains by over 20%, while it will add a fraction of a percent to global supply by getting oil to the Gulf. It’s like pouring water from a pitcher into a bathtub – the experience is likely to negatively affect the pitcher’s water supply a lot more than it will positively affect the bathtub’s.

b. The problem here is that Keystone XL will take oil from the Midwest, where refineries maximize gasoline production, to the Gulf Coast, where refineries maximize diesel production. That means less gasoline available (in this case, in the Midwest). While the Gulf Coast does produce gasoline available for export, the Midwest, East Coast and Latin America compete for those supplies – if there is less gasoline in the Midwest, it will replace that gas from the Gulf Coast, which means less gasoline for the East Coast. This creates a bidding process which ultimately results in equilibrium and higher prices.

Moreover, the higher prices in the Midwest risk putting these refineries in a similar place that East Coast refineries find themselves in – faced with higher crude oil prices, these refineries need higher product prices to maintain full capacity. Either gasoline prices go up, or refineries begin to idle. Both have happened in the east coast.
but it’s important to understand that Keystone XL isn’t expected to increase the amount of refined products Texas refineries process, it will just back out other sources. So essentially, what you’re looking at is taking about 830,000 bpd from the Midwestern and Rocky Mountain refinery districts, and throwing about 830,000 bpd

2. If you compare Midwestern refineries to Gulf Coast refineries, you’ll see that Gulf Coast refineries have slightly higher margins (or the difference between how much they sell refined products for and the amount they paid for crude oil).
Keep in mind, Gulf Coast refineries are paying the international price for crude oil – on average about $14 more per barrel than Midwestern refineries (and much of that is because of cheaper Canadian crude). Even with that discount, these refineries aren’t able to make as much money on their product as Gulf Coast refineries are, because they don’t have access to the lucrative international market for refined products. As high as U.S. gasoline prices are, in many they’re not high enough to support the international price of crude oil.

Refining margins:

Refinery acquisition costs:

3. The issue with heavier, lower quality crudes is they tend to produce larger volumes of heavier products (and diesel is heavier than gasoline). However, crude quality does not appear to account for the higher gasoline yields in the Midwest – for example, refineries in the northern Midwest are processing heavier, more sulfuric crudes than those in the Texas Gulf Coast, and yet still have significantly higher gasoline yields than Texas Gulf Coast refineries do.

ANthony SwiftMay 23 2012 11:12 AM

BS: I’d encourage you to read the report and take a look at the footnotes. Regarding what Canada produces and where it goes, take a look at footnote 19. Our analysis includes both conventional crude and tar sands. Some of Canadian production is used domestically, and additional quantities are produced on the Atlantic Coast – these volumes don’t require export pipeline capacity across the U.S. border.

Once Canadian pipelines cross the border, some may take on U.S. crude and domestic U.S. pipelines can take on Canadian crude. The cross border export capacity is the critical number – and we use Canada’s numbers to evaluate that. As for Keystone XL, an on-ramp in Bakken has been proposed, which would carry a maximum of 100k bpd (12%) of Bakken crude. I’ve written about this, but it is not relevant to the analysis of the KXL’s impact on U.S. gas prices.

Including rail transport as an alternative to pipeline transport would make the findings of this report more aggressive – excluding rail provides a more conservative view of transport capacity for Canadian crude. Even under this conservative view, there is significant excess capacity.

As a note - I welcome comments, but for the next couple weeks, I'm not going to have the capacity to personally go back and forth on these issues in them.

Michael May 23 2012 04:25 PM

Question to all... "are YOU responsible enough to know what is right and what is wrong?" This is based upon actual thought. George Costanza said it best and it applies to all those involved, "it's not a lie, if you believe it" Choose wisely people for when your dead, people still must live on this earth or whats left of it. Fact: The Ogallala Aquifer is the largest underground water source in North America. Fact: All other current pipelines run around it. Now due to reasoning such as "economic" strife, we actually would consider running this line right through the heart of this water source. Money talks, because people can't anymore!

BSMay 23 2012 07:34 PM

Michael--I suggest you look at a map of pipelines overlayed on the aquifer before you claim that pipelines run around it. And the proposed line (even before it was moved east) does/did not run through the heart of the aquifer.

I do think moving the line further from the heart of the aquifer was a good move. Even if the risk was minimal there was no reason to put it that close to the heart of the aquifer in the first place.

BSMay 23 2012 07:47 PM


You continue to claim that building Keystone XL will somehow take oil away from Midwestern refineries. There is no truth to this, and I've asked you to explain yourself more than once.

As I've pointed out before, and I'm sure you are well aware, the first Keystone system had long term contracts to ship oil to midewster refineries before the pipeline was even built. Shippers on Keystone wouldn't be allowed to redirect those barrels even if they wanted to.

Additionally, the rate of return on the first Keystone system was extremely low (and again, only built because they had the long-term contracts). Taking oil away from that system and diverting it to the Gulf Coast would basically destroy TransCanada's investment in that pipeline. No company in their right mind would spend billions on one investment knowing that it would cost them billions on another investment.

BP is spending over $1 billion to expand its Whiting refinery to process MORE Canadian crude. If they were going to loose access to Canadian crude, do you really think they would be doing that? Of course not. That investment is contingent on contracts with Canadian producers/shippers that they will be able to get the crude oil in addition to the oil already contractually committed to that region of the country.

Again, you continue to claim that the Midwest will lose their Canadian oil, but there is absolutely zero evidence to support that claim.

Please address this and then we can move on to the next topic.

BSMay 23 2012 09:29 PM

By the way, you and Simon Mui need to coordinate your efforts a little better. While you release a report that, among other things, talks about how increased oil production has helped lower gasoline prices significantly in certain parts of the US, Simon puts out an article saying increased production has not impacted gasoline prices.

Some might consider these two claims coming from the same organization to be slightly contradictory. You can't have it both ways.

Anthony SwiftMay 23 2012 10:48 PM

BS: Very quickly – Keystone XL would divert crude oil from common carrier pipelines such as Enbridge’s Lakehead system. Enbridge has said as much during TransCanada’s initial application process in Canada. Keystone XL will not take oil from Keystone I – but as we’ve discussed in detail, Western Canada exports over 2 million bpd; and only about a quarter of that is on Keystone I. Most of the rest is not under long term contract and can be diverted on a new pipeline that goes somewhere else (the Gulf) that will pay more for crude than the Midwest is.

As for BP, keep in mind that the majority of U.S. imports from Western Canada these days are tar sands. While Keystone XL will divert a significant quantity of Midwestern crude oil supplies, it won’t transport all of it. That said, the Canadian crude that stays in the Midwest will be more expensive.

I don’t think the report contradicts the reality that as the United States, we can’t drill our way to lower prices. After all, the U.S. is producing more oil than it has in years, and yet oil and gasoline prices are at sustained historic highs (even in the Midwest where prices are below international levels). We’ve been playing the drilling game, winning it when it comes to production, and still losing the war on prices. I don’t think that argument is undermined by saying that reducing the amount of gasoline we produce from the oil our refineries process is going to have an impact on gasoline prices. And KXL will do that by shifting crude from the Midwest to the Gulf.

BSMay 23 2012 10:56 PM

I see what you're saying, but Keystone XL isn't built yet. Canadian production is expected to grow +300Mbpd this year alone, and more next year. By the time XL is built, what it carries will be roughly equivalent to the increase in production over the next couple of years.

I'm not arguing on the price issue. Of course debottlenecking the supply of Canadian oil will increase the price of that oil. Prices in the Midwest will go up, but prices globally will go down due to the increase in supply. And the increase in the price of Canadian crude will drive even more production increases, which will keep all supply routes supplied.

I suspect that last point is what the NRDC is really worried about. That higher prices will lead to even more production.

BSMay 23 2012 10:58 PM

"I don’t think the report contradicts the reality that as the United States, we can’t drill our way to lower prices."

I've commented on this in Simon's blog. Rather than repeat myself, I'd invite you to read what I've written there.

There is absolutely no doubt that expanded domestic and Canadian production has resulted in prices being lower than what they otherwise would have been.

Also, you may have noticed that the price of WTI is down to $90/bbl today. One of the principle reasons for that is an excess supply. There are other reasons, too, but that's the one that is relevant to the discussion.

Mike H.May 24 2012 11:12 PM

Looking at the preliminary NTSB report from the 2010 Enbridge crude oil pipeline rupture in Mashall Mich., which was carrying tar sands crude at the time, there was confusion if what pipeline Controllers thought what was going on was a phase separation, or a leak. Yes, dliutents don't stay well mixed with tar sands crude. So, will more expensive dillutents be needed to prevent that?

There's also still a LOT or crude left from that spill. Conventional crude clean up techniques doesn't work too good, it seems.

BSMay 25 2012 08:48 AM

Mike--I don't know all the details, but the spill was likely preventable with better training. Two liquids separating into two phases doesn't account for less oil coming out of the pipe at the terminal than what is going into it at the origin.

There really is no such thing as a more expensive diluent that would be feasible to use.

Comments are closed for this post.


Switchboard is the staff blog of the Natural Resources Defense Council, the nation’s most effective environmental group. For more about our work, including in-depth policy documents, action alerts and ways you can contribute, visit

Feeds: Anthony Swift’s blog

Feeds: Stay Plugged In