As Statistics Canada continues to roll-out the results from the National Household survey, I seem to become involved in arguments at least once a week as to the importance of sample selection in survey data. This week, my argument was with IPSOS CEO Darrell Bricker – someone who should know a lot about statistics. In particular, Mr. Bricker should know that you can’t solve a sample selection problem with an increased sample size, and I actually think he does. I think the issue is that he’s thinking about practical polling issues with respect to sampling, not about statistical issues with respect to selected samples. Statistics Canada differentiates between sampling error and non-sampling errors, and I think that’s where our key difference lies. Let me see if I can explain this, and hopefully Mr. Bricker will respond and let me know if I am on the right track.
By Andrew on September 11, 2013
My latest at Macleans digs into TransCanada’s analysis on crude cost savings for eastern refiners from Energy East. Check it out here.
By Andrew on September 6, 2013
One of the things which changed during my year in Ottawa is that I became a year-round bike commuter. I’ve decided to try to keep this up in Edmonton, despite some (potentially important) differences in climate and snow clearing between the two cities.
One of the factors in my initial decision to ride to work was savings in terms of the cost of gas and parking, which made up for most of the additional cost equipment required to bike through the winter in just one year. At Environment Canada, the cost of a monthly parking pass was $170, compared to a daily rate of $9. It was easy to figure out that, as long as I could expect to ride to work more than a couple of days per month, I would be ahead on parking costs not to mention fitness. In January, I drove to work 5 times, which was the most all year, so my worst month saw me save $125 on parking, plus the savings in fuel. So, each morning I had a no-bike tax of $9, but the average cost to me each month was actually negative.
This pricing relationship is very different at U of A as a monthly pass costs only $96, while daily parking is $14. I’ll have to ride to work at least 14 days out of 20 to make it pay to not get a parking pass. However, if I don’t buy the pass, my marginal incentive (my no-bike tax) will go up to $14/d making it much less tempting to drive to work on the really cold days. It’s possible that, in the cold months of the year, I’ll lose money by not buying a parking pass but I’ll bike more if I don’t have one.
I might be out a few dollars, but I’ll be in better shape in the Spring. Sometimes, economists do think about things other than money.
By Andrew on September 3, 2013
By Andrew on August 30, 2013
There are 4 major milestones in a professor’s career – you get your first job as Assistant Professor, you get tenure, then (or possibly jointly) get promoted to Associate Professor, and finally to (full) Professor. Outside of that formal structure, there are separate rewards that you might earn – Professorships, Chairs–including those endowed by donors or funded through a government granting agency–fellowships, etc. This year, I was awarded one of these – the Alberta School of Business Enbridge Professorship in Energy Policy. I’m grateful for the recognition from the school and grateful for the support of Enbridge, as well as for that of our many other donors. The University of Alberta’s Donation Acceptance Policy states that, “philanthropic support is an important element in advancing research and education,” and there are many donors who make it possible to do the work we do at the Alberta School of Business. I am aware that many of you will have questions about this professorship and how it influences my work and responsibilities. I hope that this post will answer them.
I am committed to transparency and disclosure with respect to potential conflicts of interest. I have posted my own Conflict of Interest Disclosure on this blog and kept it regularly updated since early 2012. I began this in response to the American Economics Association’s adoption of new guidelines for economists in terms of conflict of interest management, also in 2012. The link to my conflict of interest disclosure is on every page, at the top of every post. If you want to know anything about me, from who has paid me to speak to who has given me hockey tickets to what stocks I am holding today, you can find out with a click and a scroll. If there’s something you’d like to know that is not there and is germane to my public engagement or research work, let me know and I’ll add it to the list. This post will have a permanent link from that page as well.
I have structured the remainder of this post as a Q&A.
What is the Enbridge Professorship? The Enbridge Professorship is an award supported by a donation from Enbridge to the Alberta School of Business. The award is given by the Alberta School of Business to a faculty member working in the field of energy policy. The award provides financial resources, which may be used to support research or taken as salary. The total value of the award is less than 10% of my annual salary. The award does not change my conditions of employment at the university as stipulated by the Faculty Agreement (PDF), nor does it change my rank. I remain an Associate Professor with tenure.
Does this mean that Enbridge is paying part of your salary? Yes and no. The Professorship is an appointment provided by the Alberta School of Business funded through a donation from Enbridge. My salary is paid by the University of Alberta, but I can elect on an annual basis to receive some or all of the Professorship resources as a stipend, which would be funded by the Enbridge donation. As specified above, the share of my annual earnings which would be funded by the Enbridge Professorship would be a maximum of less than 10%. Based on my elections for this academic year, the share will be less than 5%.
Can the position be revoked at Enbridge’s request? No. The professorship and accompanying resources are provided by the Alberta School of Business, and so such a decision would be at the School’s discretion, not Enbridge’s. The University of Alberta has specific guidelines that detail the relationship between a donor and the university, protecting scholarly freedom and integrity, which you can see here.
Are you required to be supportive of Enbridge projects, actions, statements? No. The Faculty Agreement is very specific as to the protection of academic freedom. Article 2 of the agreement states that the principles of academic freedom include, “the right to examine, to question, to teach, to learn, to investigate, to speculate, to comment, to criticize without deference to prescribed doctrine.” This is not superseded by the awarding of a sponsored professorship. The University of Alberta’s Donation Acceptance Policy is very clear in this regard, and states that, “the University values and will protect its integrity, autonomy, and academic freedom, and does not accept donations when a condition of such acceptance would compromise these fundamental principles.”
I’ve been supportive of Enbridge positions in some cases in the past, and not so at other times. I expect that will continue. I have received wonderful support in my classes from Enbridge over the years, as well as from many of their competitors and opponents. I expect that to continue as well.
Does the award specify subjects that you should or should not address in your research? No. The award is not a research grant and so is not tied to a specific project, nor are there expectations in terms of specific research questions to be addressed by the holder of the professorship. I expect to continue as before – working on what interests me, and trying to publish articles in the top peer-reviewed journals in my field, as well as maintaining this blog and other public platforms. In accordance with the American Economics Association guidelines, I will disclose the general research support provided through the Enbridge Professorship on research papers submitted for publication, and on my conflict of interest disclosure, which is readily available on this blog. The award places no restrictions on research topics or any other academic activity, as explained above.
Do you expect me to believe you aren’t conflicted by this position? No. There is no question that the debate over energy is both very important and often heated in Canada. As a result, in accepting this position, I was and remain well aware that people will assume that I am conflicted by this position and will view my writing and commentary through that lens. That’s fine – please read and judge for yourselves.
My promise to myself on accepting this position was that if I were to come to feel that my academic freedom had suffered as a result, I’d give up the professorship. Academic freedom is the first substantive clause in the Faculty Agreement for a reason: it’s crucial to have that freedom in order to contribute positively to the public understanding of important issues including energy. I am not prepared to give that up.
Did Enbridge approve this post? No. A copy was provided in advance to the appropriate contact within Enbridge as a courtesy.
I hope that this has addressed any concerns you may have and that you will continue reading my work here and elsewhere.
By Andrew on August 8, 2013
It’s been a week since TransCanada announced that it had secured sufficient commercial commitments and would be proceeding with the Energy East project. Their announcement included a few surprises – a larger-than-expected capacity of 1.1 million barrels per day, and a $300 million marine terminal in Saint John. What follows is a Q&A addressing some of the fact and fiction that’s been tossed around this week.
Is this pipeline a done deal?
Not even close. Securing commercial commitments is an important step, but there remains a long process ahead. The project will still face at least National Energy Board approval, but more likely a federal joint review panel including both the Canadian Environmental Assessment Agency and the National Energy Board, similar to the proceedings currently underway for the Northern Gateway Pipeline. There will also be consultations with First Nations, negotiations with landowners with respect to the new sections of pipeline, as well as dealings with the 6 provincial governments affected by this project among other requirements. Provinces do not have to approve energy pipelines in order for them to be granted a Certificate of Public Convenience and Necessity by the National Energy Board, but the process will be much smoother for the proponent if the respective provinces are on side.
Will Energy East lead to lower gas prices?
The answer to this comes in two parts – first you need to ask what Energy East will do to crude costs for refineries across Canada, and then you need to ask whether such a change will alter gasoline prices at the pump.
Energy East will serve as a link between Eastern Canadian refineries and the western crude oil market, where crude oil had been discounted significantly since mid-2010 until these price differentials converged rapidly over the last couple of months. By providing this link, the pipeline will alter future differentials from what they would otherwise be. If we assume that the marginal barrel out of Western Canada is still moving by rail, then you would expect western Canadian oil to be priced at approximately $12-15 below prevailing prices at the coast, reflecting the cost of rail transport. In such a scenario, a refiner who owned firm shipping capacity on Energy East could purchase oil at Edmonton and deliver it to Montreal, Quebec, or Saint John at a lower cost than that at which they would otherwise be able to obtain oil. Refiners without their own firm shipping capacity on Energy East (or Enbridge’s Line 9) will continue to pay world prices for oil, regardless of whether that oil is Canadian or imported. If the marginal barrel is moving by pipeline, the difference between oil in Alberta and oil on the east coast is likely to be comparable to the shipping tolls on Energy East, and so running Canadian oil would not save refiners any money at all relative to running imported crude.
There is some evidence that recent crude oil discounts in Western Canada have been partially passed through to consumers through lower gas prices. Statistics Canada reported in The Daily in November that, “gasoline prices have increased at a slightly faster pace in the central and eastern provinces than in the west, resulting in a spread between some provincial gasoline indices…associated with the dual crude oil market in Canada and the recent price differential between crude oil benchmarks.” The image below shows that this pattern has continued for the last 6 months since the Statistics Canada report.
Source: Statistics Canada
However, you need to keep in mind that we are not talking about a systematic lowering of crude oil costs in eastern North America – we are talking about an increase in crude costs in Western Canada, combined with a potential small decrease in costs for some eastern refineries. As a result, you may see a small increase in refining margins for those refiners with firm shipping capacity on Energy East, along with a decrease in refining margins for in-land refineries in Alberta. If you’re going to see a change at the pump as a result of this project, I’d expect to see an increase in Western Canada rather than a decrease in Eastern Canada.
Will Energy East increase oil sands development?
As with the question above, this question should really be followed with an alternative scenario. If you evaluate Energy East against an alternative scenario where no other pipelines out of Alberta are ever built and no existing pipes are expanded, you could expect a marginal impact on future oil sands development. I say marginal because you simply can’t look at it as enabling 1.1 million barrels per day of new production which would otherwise not occur – you have to look at the impact on profitability between the two scenarios and ask whether that impact is sufficient to have an impact on the pace of development. For a bitumen project, the difference between pipeline and rail is small, once diluent transport is factored in. In order to ship a barrel of bitumen by pipeline, you’d need to ship 1.4 barrels of diluted bitumen, so if your pipeline heavy oil toll is $7/bbl, it is going to cost you $9.80 to get a barrel of bitumen from Alberta to Saint John. You can ship bitumen by rail with little or no diluent added, so the rail vs. pipe trade-off is likely to be only $3-5 lower netback per barrel. That type of decrease in expected profit certainly matters, but it’s not likely to reduce oil sands production by 1.1 million barrels per day.
You might argue that the rail system will, at some point, reach a capacity constraint. That’s possible, but unlikely. Despite all the attention to oil-by-rail increases, crude and fuel oil still only accounted for less than 5% of Canadian rail car loadings in May, 2013. Nickel, iron-ore, coal, and potash each accounted for more rail car loadings than crude oil. Access to pipelines matters, but given the alternative of oil-by-rail, the profitability impact is not as large as some have made it out to be and that’s what will drive development.
Should we have built this pipeline a long time ago?
There’s a strong nationalist element which seems to think we’d be somehow better off had we not been importing oil on the east coast while exporting oil from western Canada to the US Midwest. That sentiment ignores the market reality for much of the last 25 years. Here’s an admittedly simple simulation using crude oil prices since 1987. Assume you could have built a pipeline and put it in service from Alberta to an eastern refinery center, with a pipeline toll of $7/bbl in today’s dollars, adjusted for inflation (the toll would have been $3.89/bbl in 1987). Assume that your alternative project would have been the Keystone Pipeline to Cushing, with tolls equivalent to today’s tolls on that line to Cushing, OK, again adjusted for inflation. Now, here’s your choice – you can either ship your oil to Cushing, and sell it there and use the revenues to buy oil on the east coast, or you can ship your oil all the way to the east coast, and sell it (assumed price of Brent +$1.50). (the simple part of this simulation is that I’ve assumed that prices in Cushing and in eastern Canada are not affected by your choice). The Figure below shows how much you would have lost on every barrel if you chose the Canadian option.
From 1987 to 2010, you’d have been better off in every single month making the trade we have been making – exporting to the US Midwest and importing on the east coast. In fact, the from May 1987 to May 2013, the average loss from shipping 1 million barrels per day of oil east rather than south would have been about $250 million per month in today’s dollars. If you assume you could have moved it east for free, you’d have still lost money, although not much of it.
The oil market in North America has certainly changed, and the expectation is that the marginal barrel will be moving to the coasts, so oil in the mid-continent will no longer trade at a premium as it has. As a result, the economics of Energy East are likely better today than they would have been at almost any point in the last couple of decades. It’s certainly the case that perfect foresight would have led to more pipelines being built to the coasts instead of to the Midwest in the late 2000s, but not before that.