15

Checking the math on Energy East

Our in-house economist finds higher minimum savings and lower maximum savings than recent reports suggest


 

Daniel Acker / Bloomberg / Getty Images

One of the main selling points of TransCanada’s Energy East Pipeline is that it presents a potential win-win whereby eastern refiners would be able to pay less for crude than they otherwise would, while paying more for Alberta crude than Alberta producers would otherwise receive. Yesterday, TransCanada announced the results from a study on the economic benefits of Energy East. One of the lines that caught my eye in that detailed report, authored by Deloitte, (PDF) was a statement that the pipeline would provide, “a supply of domestic crude oil sources for eastern refineries, which is expected to result in an annual feedstock cost savings of between $1.55 and $11.49 per barrel.” Now, $11.49 per barrel seemed awfully high, so I dug into the assumptions a little. I then re-estimated the potential savings, using Deloitte’s assumptions, and found $3 to 6.04 per barrel in savings for eastern refiners.

The Deloitte analysis lays out two scenarios. In the first one, the price eastern refiners pay for oil transported on Energy East is based on Brent, a north sea oil price benchmark. In the second scenario, the price they pay for Alberta oil is linked to how much it would cost to ship that oil to Europe via the Gulf Coast. These are defensible, though not the only, potential outcomes of the North American oil market with Energy East in service.

Eastern refineries currently use crude for which they pay world (or Brent) prices, and also have to pay the transportation costs (assumed in the analysis to be $1.50 per barrel to Saint John and $2.80 per barrel to Quebec) to get that oil to their refineries.* In the first scenario explored by Deloitte, the Energy East pipeline would, effectively, reverse this relationship. Oil would be flowing outward to the coast, so the price of a barrel of oil at the port of Saint John would be Brent less the price of transportation to Europe, representing a savings to refiners that is equal to double the current transportation cost. This suggests that the price at St John would go from Brent + $1.50 to Brent – $1.50, a savings of $3 per barrel relative to today’s costs. Similarly, in Quebec, you would see oil being worth Saint John prices net the toll to ship to Saint John, so Brent prices – $2.75 per barrel using the assumptions in the Deloitte analysis, a savings of $5.35 per barrel relative to current pricing. For this scenario, it appears the TransCanada’s report may have under-estimated potential crude savings (assuming the marginal barrel is moving out by export), perhaps by missing the double-impact of the crude transport cost premium being reversed.

Just so you are clear that my math works, you can make this argument from the producers’ perspective as well. If a producer can ship oil from Alberta to Saint John for $8.25 per barrel, and then to Europe for $1.50 more, they will not accept a price less than Brent – $9.75 per barrel from a domestic refiner as long as exports are permitted. So, let’s assume that the refiner buys oil in Alberta at Brent – $9.75, and pays $8.25 to ship the oil to Saint John — they’ve obtained oil at Brent prices less $1.50 per barrel, a cost savings of $3 per barrel over their previous crude prices.

Looking at the high-end of savings estimate, the Deloitte analysis again relies on assumptions as to the marginal barrel leaving Alberta. The analysis lays out a scenario where the best alternative for an Alberta producer is to ship via other pipelines to the Gulf Coast and on to Europe, and they calculate that, “the Quebec refinery savings would be equivalent to the cost to ship crude oil from western Canada to alternative markets.” The Deloitte analysis calculates this savings at $10.24 per barrel in Quebec and $11.49 per barrel in Saint John.

The calculation leading to these large potential savings seems flawed to me. Yes, it’s true that if the marginal barrel is leaving Hardisty, Alta., via Cushing and the Gulf Coast, we would expect, using  their toll assumptions, for the price of oil at Hardisty to be Brent – $10.24 per barrel. However, in order to benefit from these depressed prices via Energy East, refiners with firm service access on the pipeline would still have to pay pipeline tolls (producers aren’t going to sell at an even larger discount to cover tolls to eastern markets). If this shipping toll is $7 per barrel to Quebec, then you would see their cost of Alberta oil of Hardisty prices + $7 per barrel, or Brent prices minus $3.24 per barrel. This would amount to savings of $6.04 per barrel relative to their assumed current option of paying Brent + $2.80.

For Saint John, the high-side calculation is similar.  Using a pipeline toll of $8.25, and a Hardisty oil price of Brent less 10.24 per barrel, the price of Alberta oil delivered to Saint John via Energy East would be Brent less $1.99 per barrel, a savings of $3.49 per barrel relative to their current pricing of Brent + $1.50.

So, using the assumptions laid out in the report, I’d calculate potential savings of $5.35 to $6.04 per barrel for a Quebec refinery and and of $3 to $3.49 per barrel in Saint John.  These results have higher minimum savings and lower maximum savings than those reported by TransCanada and Deloitte.

*As an aside, it previous writings on the issue, I’ve made the assumption that eastern refineries are paying Brent prices and have ignored the transportation cost premium paid relative to Brent.  As the analysis above shows, this transportation cost premium, and its potential reversal, plays a very important role in determining the potential cost savings from the pipeline project, so I am happy to correct that previous oversight.

 


 

Checking the math on Energy East

  1. So, given that you now calculate “potential savings of $5.35 to $6.04 per barrel for a Quebec refinery and and of $3 to $3.49 per barrel in Saint John” can we conclude that security of supply is a bonus, or an “externality” as you guys like to say?

    • You might consider any security value a positive externality, yes. To fit the standard definition, if the refiners receive a benefit for which they do not compensate producers or if Canadians receive a benefit for which they don’t pay either refiners or producers, those would be externalities.

  2. Bunk, just the propaganda here.

    Reality is the crooked do not want to see western resources go for fair market value and do not want the west to have Pacific coast access. Sort of like NEP II, closed market access for lower prices to screw the west. The reason why I support the concept of the Republic of Western Canada. Time we ditch colonialism.

    • True. Just pass a county of Alberta resolution that the Pacific Ocean starts at the BC border. Yeah, that’s the ticket. Landlocked problem solved.

  3. Andrew, thanks for the great article. I was wondering if you could explain the following statement “Oil would be flowing outward to the coast, so the price of a barrel of oil at the port of Saint John would be Brent less the price of transportation to Europe,”

    Why would eastern refineries have to pay the transportation costs in the current scenario, but then the buyer in this alternate scenario not have to do so?

    • Good question John. There are two assumptions buried in this. First, the assumption is that the current supply costs are best proxied by Brent prices, plus shipping costs. I could have done the same analysis with Middle East prices, but then it would have had a larger transport cost premium. When you are trying to get barrels TO your refinery, you have to pay enough to both get the oil and get it there, hence Brent + shipping. If the flow is reversed, another assumption comes in – that the best alternative for a Canadian shipper would be to sell the barrel in Europe (to make a Brent trade, basically). If that’s true, then if a refiner offered you Brent minus the costs to get it there, you’d be indifferent between selling it to the refiner or shipping it to Europe. Hence, the direction of “flow” changes the incidence of shipping costs.

  4. Dr Leach, is it reasonable to assume that oil exported from NB would go all the way back to a Brent+0 market, or would it more likely reach another Brent+1.5 (or higher) destination for the $1.50 transportation cost, such as elsewhere in PADD I?

    • Another good question. It’s certainly possible that it could reach another Brent + 1.5 market, and at less than 1.5 shipping costs. That would give higher value to shippers, less savings to eastern refiners. I’ll make sure to use this as one of the cases I’ll consider in an upcoming piece. In this piece, I was using the assumptions in the Deloitte report for the most part, so I didn’t go that far. Thanks.

      • To be fair, this was covered in the Deloitte report (although it appears they forgot to deduct the p/l toll to eastern Canada as you have pointed out):

        Quebec refinery savings (project)

        = Cost to ship crude oil from western Canada to alternative markets

        = (Toll from Hardisty, AB to Cushing, OK) + (toll from Cushing, OK to U.S. Gulf Coast) + (shipping costs to farthest alternative market, Europe)
        .
        = $6.05/barrel + $0.44/barrel + $3.75/barrel = $10.24/barrel
        .
        St. John refineries could potentially receive an additional $1.25/barrel in savings, or $11.49/barrel total, as shipping down the St. Lawrence River would not be required.
        .
        It was also covered indirectly in the Scotiabank Commodity Price Index July 30, 2013
        .
        Refiners in India have shown considerable interest in importing Alberta blended bitumen. Estimated tanker charges from Québec City & Saint John to the west coast of India average a mere US$4.20 per barrel in a Suezmax vessel. A marine terminal at Saint John would be ice-free year round and could accommodate VLCCs of 350,000 DWT, cutting tanker costs to India to only US$3 per barrel.

        • Yes, I think Deloitte got this one at least partly right. They added the 1.5 to get the oil to PADD I or Europe on the low side as well. Strange that they forgot so many other significant costs. I haven’t known Deloitte to have a major energy practice. If they want to build one, they might hire Dr Leach!

          • Yeah, I’m not so sure of that. I’ve seen mention of them focusing on O&G operations (reducing costs) from their Calgary practice. Often, they hire individuals with prior industry experience (Oil and Gas) in addition to business (MBA).

            It was a silly oversight in the report. AL picked it up quickly (forgetting to deduct east p/l tolls) on twitter, though, to his credit.

          • Hopefully they’ll keep hiring my students.

        • Actually, nothing in that quote covers Alex’s point. If you could ship to another “premium to Brent” market – eastern US, for example – from Saint John, for less than the cost of shipping to Europe, you would raise the price at Saint John to closer to (or potentially above) Brent, This would decrease potential savings below the ranges considered in the Deloitte analysis or my rebuttal. The report states that shipping costs to India are $3-$4.20, but what matters is potential pricing in India that is favourable to by enough to cover the most of the shipping costs. If so, that could change the price at Saint John (i.e. if crude traded at a $5 premium to Brent in India, you could effectively net Brent + either $0.80 or $2 depending on your vessel choice at Saint John, raising the cost of crude to domestic refiners relative to the Deloitte case). In all of their analysis of crude costs, as you point out Derek, they implicitly assume that “best alternative market” was Europe although they label it as farthest.

          • Well, I tend to agree with the Deloitte report on the lower bound of $1.50. Could be lower, I guess, if you can load/unload and ship for less than $1.50. Haven’t seen any numbers suggesting you could. $1.30 to ship from St. John up the St Lawrence to Quebec in the Deloitte report. And in the Scotiabank report, $1.50 from St John to PADD I (northeast US).

            If one accepts that $1.50 is the cost to ship from Scotland for Brent to NB (and let’s keep in mind, the unloading and unloading costs can be assumed fixed, mileage/time at sea is variable) then hard to imagine there are many places cheaper than $1.50. Keeping in mind Brent is a benchmark and that other crudes are priced relative to it, taking into consideration distance and quality. Naturally, one would not ship to Scotland, and unload there – to be reloaded and sent elsewhere.

            I sort of look at these alternatives (India for example) as worst case scenarios. If it costs $3-$4.20 /barrel from St John, presumably similar numbers from Scotland (ocean distance/time seems about equivalent). Doesn’t this set the upper price limit for Canadian crude (worst/best case scenario depending upon your perspective)? If India can get it cheaper (say it got it from Saudi Arabia corrected for grade and shorter shipping distance) then the upper limit would be lower.

            In any event, we’re probably arguing the same thing. Apart from missing out on the eastern p/l tolls, I thought the Deloitte report was pretty good as a high level study. To be refined, no doubt as scope and costs firmed up.

          • I agree that $1.50 is a reasonable limit, although again it matters to where. If you are shipping at $1.50/bbl to PADD I, which would normally be paying Brent plus $1.50, then your value at port in Saint John is equivalent to Brent.

Sign in to comment.