Would Enbridge's Line 9 really save Quebec refineries $23 billion? - Macleans.ca

Would Enbridge’s Line 9 really save Quebec refineries $23 billion?

Andrew Leach checks the math on a very big number in the pipeline hearing


This article appeared first on CanadianBusiness.com

The National Energy Board is about to wrap-up hearings on Enbridge’s proposal to reverse the flow of one section of Line 9, an oil pipeline which now flows from Montreal, Que. to Westover, Ont. The west-to-east reversal, along with a 25% expansion in shipping capacity, would mean the pipeline could ship up to 300,000 barrels per day of crude from Sarnia, Ont. to Montreal. The project, Enbridge estimates, will save Quebec refineries $23 billion over the next 30 years. Yes, billions.

I decided to check some of the numbers on this project, as I did in a similar math-check on TransCanada’s Energy East project. Much as with Energy East, the economic case for the reversal of Line 9 is that it is expected to be cheaper for Canadian refineries to use North American crude as compared to crude imported from overseas for the foreseeable future. That wasn’t the case until recently. As pipeline infrastructure and crude oil markets adjusts to new U.S. and Canadian production and declining North American demand, more oil is moving from the mid-continent to the coast. Oil pipelines in the continent used to flow primarily to points in the Midwest, but as that area has seen increases in domestic supply as well as increased imports from Canada, it is now over-supplied with oil. As shipping infrastructure is re-aligned to ship oil out of the Great Lakes region, where oil used to trade at a premium to world prices because it had to be shipped-in from coastal markets, oil in the mid-continent will likely trade at a discount for the foreseeable future. Over time, we should expect the discount on mid-continent crude to be $3-5 per barrel plus inflation compared to world market prices, since that’s about what it would cost to ship a barrel south to the Gulf Coast. Refineries using mid-continent crude, then, are likely in for some savings. The question is whether these savings can truly amount to something as astronomical as $23 billion.

So, how does Enbridge get $23 billion? They state in one of their regulatory filings that, “the (crude cost) savings are estimated to amount to approximately $780 million annually over 30 years, based on forecast oil price differentials, throughput of 250,000 barrels per day, and deliveries divided equally between Montréal and Québec City.” Another way to look at it is average cost savings of $8.54 per barrel over the next 30 years. (Note here that Enbridge is speaking about light oil, not bitumen. Quebec refineries are geared toward processing light crude and would have to embark in a costly retooling in order to handle heavy crude from the oil sands.)

Instinctively, I was suspicious of a number as big as $23 billion. A few back-of-the-envelope calculations suggest Enbridge’s figure is at least a modest overestimation.

Refineries in Montreal and Quebec City currently rely on imported crude for which we normally use UK Brent crude as a price benchmark. On top of Brent prices, they generally have to pay a shipping charge of about $2 per barrel. For crude from the mid-continent, instead, people use a different price benchmark called West Texas Intermediate, or WTI. You could start with a simple calculation, and assume that we’d need to see WTI  price discounts to Brent crude large enough to make up $8.54 per barrel in savings, plus the cost of shipping the crude from Cushing, Oklahoma (where WTI contracts are settled) to Montreal, minus the $2 in shipping costs refiners would have otherwise had to pay on Brent-based imports. To make Enbridge’s math worth with those assumptions, you’d need to expect WTI crude to be about $14 per barrel cheaper than Brent.

That is a lot. The chart below shows the WTI-to-Brent price discount for the next 30 years as forecasts by the U.S. Energy Information Administration (15-04-2013), Sproule Associates (09-30-13) and GLJ Petroleum Consultants (10-01-13). All projections suggest that discounts will decrease from today’s levels but remain above zero. Still, they vary widely — and no one is predicting differentials of $14 per barrel.

Forecast discount of West Texas Intermediate (WTI) crude oil relative to world prices from the Energy Information Administration (EIA), Sproule Associates, and GLJ Petroleum Consultants

Does this mean Enbridge’s math doesn’t work? To understand how Line 9 might allow refiners to access lower cost crude, you need to look at oil pricing within North America in yet more detail. That requires a couple of assumptions. I am going to assume that the mid-continent crude market will see the lowest prices around the tip of Lake Michigan, from which oil is either shipped south to the oil terminal of Cushing, Oklahoma, where WTI contracts are settled, or north to Canada. If these market dynamics play out, the price of crude at the tip of Lake Michigan should be roughly equal to the WTI prices minus $3 per barrel. It’s this discount, added to the premium to Brent currently paid by Quebec refiners, that makes the savings claimed by Enbridge plausible. Oil shipped to Montreal would still face pipeline tolls to the U.S. border and onward to Montreal of about $4.50 per barrel, which means a refinery in Montreal is likely going to be looking at costs of crude of WTI plus about $1.50 per barrel with the pipeline reversal in place. So, even if WTI traded at par with Brent, refiners would still be ahead by a bit. In my scenario, the oil price differentials on the graph above are gravy.

What do the crude price forecasts used to build the graph above mean for the economics of Line 9 if my assumptions above hold? The outcomes show a broad range, but none of the three get you to Enbridge’s $23 billion figure. Using the EIA forecast, which has the narrowest future differentials, we’d expect to see crude cost savings of $10.8 billion for Quebec refineries over 30 years. The GLJ forecast is better news, with $12.4 billion in decreased crude costs. The Sproule forecast would see refineries saving $20.7 billion over 30 years — close to the figure reported by Enbridge, though still a bit lower. A reasonable quibble with both my numbers and those provided by Enbridge is that they are reported in nominal terms, simply adding up dollars over time. A net present value calculation, with a 10% rate of discount commonly used in the oil and gas industry, would lower the value of my savings calculations to $3-5 billion over 30 years. Still in the billions, just not $23 billion.

Disclosure: Readers should be aware that the author holds the Enbridge Professorship at the Alberta School of Business, University of Alberta. You can read what that means here. This post received no input from Enbridge, nor were they made aware of it being published in advance with the exception of posts made on the author’s Twitter account here and here.


Would Enbridge’s Line 9 really save Quebec refineries $23 billion?

  1. Good final disclosure. More is always better than less.

  2. Do they really expect this line to hold up for over 70 years?

    • estimates are over 30 years.

      • Isn’t the line already over 30 years old?

        • The line will last that long, but not necessarily all of the individual pipe. Not my area, but you’ll see significant maintenance on all pipelines, including replacement of sections as needed.

  3. sadly, it is at the point where what Enbridge says is guaranteed to be suspect, and more than occasionally one could say ‘misleading’. Doing business usually implies honesty and verifiability. And if these elements are absent a purchaser will usually cease business with that company.
    So long Enbridge. If you were more open and honest you would have had a good shot at Northern gateway and 9B, but the cat is out of the bag and you are out of business with Canadians.

  4. Prof. Leach, has there been any analysis as to the positives that QC and Canada as a whole would receive from what I presume would be increase tax take from increased profits at the refineries?

    • Not directly. The analysis which Enbridge has presented, so far as I am aware, specifically excludes the tax implications of refinery cost savings.

  5. Now consider possibility of rupture. 1 billion set aside for insurance will not go far in an area like down town Toronto. Prpoetry values are high. What if the spill shutdown down York University, the Finch subway where it is currently overtook an industry in close proximity? Could get expensive. Now consider if we had another Hurricane Hazel or tornado or earth quake. We could have multiple ruptures. The fact is, this pipe is only buried three feet down (1 meter.) It has stress fractures all along it and Enbridge already identified multiple crack areas that will rupture at 740 psi or lower. They don’t want to do hydrostatic testing saying it could damage the pipe but the want to run dilbit at 750 to 1000psi. That is scary. The PE tape is the same as Marshall Michigan’s line 6. One billion clean up and counting. Both pipes: the tape is not up to code. It is too costly a gamble.

  6. ‘Instinctively, I was suspicious….’ If only media were as suspicious over environmentalist claims.

      • Over the years I have heard, among other claims:
        ‘The tar(sic)sands are open pit mines the size of Florida (in Britain they say England)…They are twice the size of Florida… They are twice the size of Florida and growing.’ The truth is the deposit is smaller than Florida, only 2.5% is minable using current technology and they unfortunately aren’t growing as far as we know. But it does seem to upset some people despite the Bakken being larger, by some accounts 4 times the size. So is the Permian Basin. And the sites are reclaimed. Unlike Vancouver where they took the best farmland with the best climate for growing and put a city on it. In fact most of our cities are on better land than the oilsands and that land is not planned to be reclaimed.
        ‘They use tremendous amounts of water.’ I thought 3.4 bbls per bbl of bitumen was a lot too. But considering ethanol uses more (as does solar) it puts things into perspective especially considering the water is either a small part of the Mackenzie system destined for the ocean with few competing uses or is produced ground water with no other use. It’s certainly less than the losses due to evaporation from Lake Mead during the production of the equivalent amount of energy from Hoover Dam. 800 times less.
        ‘It’s killing the Indians.’ The truth is bitumen has been entering the Athabasca River through erosion for eons. The health of oilsands workers (including natives) and Ft MacMurray residents isn’t known to be adversely affected by development. The health of Indians throughout the country isn’t as good as the rest of Canada to start with as they have a life expectancy of 10% less on average. But this is related partly to lifestyles connected to poverty. The oilsands, like other developments, offer employment that can alleviate these problems.
        ‘More CO2 is produced than with ‘conventional’ oil.’ This is at least a half truth. But the full truth would have to include acknowledgement that peak oil has been passed as far as conventional sources go. We’re now going for non conventional such as heavy, oilsands, shale etc. The heavy oils of California, Venezuela and Mexico are produced with similar technology as our insitu and produce the same, or more, CO2 as ours would. Nigerian light and medium is produced with flared associated gas. Bakken also has a higher CO2 footprint as a result of flaring.

        • I’ve written on many of these issues, in particular relative GHG footprints, on my blog at Andrewleach.ca. unfortunately, both “sides” in the oil sands debate are guilty of taking statistics and applying them out of context. Your example of deposit area vs. surface distirbance is a good example, as is the frequent omission of the word “intensity” when people talk about oil sands emissions declining since 1990. I don’t think it’s fair to say there has been a systemic bias in calling out industry figures. I’d argue that many job and economic impact figures are taken at face value from industry in the same way as impact on environment may be taken at close to face value from environmental groups.