Economic analysis

What we learned from Keystone XL

The debate around Keystone distracted us from a sensible discussion about environmental policies and the economic impact of pipelines
Trevor Tombe
Keystone Application Pause 20151102
A yard in Gascoyne, ND., which has hundreds of kilometres of pipes stacked inside it that are supposed to go into the Keystone XL pipeline, should it ever be approved are shown shown on Wednesday April 22, 2015. In a surprise move, the company behind the controversy-plagued Keystone XL pipeline has asked the U.S. government to temporarily suspend its application. That request from TransCanada Corp. adds a new wrinkle to one of the biggest Canada-U.S. political irritants of recent years, involving a pipeline from Alberta to Texas. THE CANADIAN PRESS/Alex Panetta

The seven-year regulatory journey of the Keystone XL pipeline is now over. On Nov. 6, the State Department decided, and President Obama agreed, that KXL would not serve the U.S. national interest.

While the symbolism is huge, the effect of this decision on the economy or the environment is not. On this point, the President nailed it:

For years, the Keystone pipeline has occupied … an over-inflated role in our political discourse. It became a symbol too often used as a campaign cudgel by both parties… This pipeline would neither be a silver bullet for the economy as was promised by some; nor the express lane to climate disaster proclaimed by others.

It’s hard to disagree with this assessment, whatever one’s views of Keystone are. The entire episode will no doubt be remembered as one of the least productive public policy debates in recent memory. It wasn’t about jobs or the climate, as Christie Aschwanden of FiveThirtyEight put it, it was politics. On the plus side, it provides a useful teachable moment to clarify some of the common environmental and economic claims one often hears.

Do pipelines export jobs? Where’s the value in exporting raw unprocessed resources? Is opposition to pipelines sensible environmental policy? There are many other pipeline projects in the works – Energy East, Trans-Mountain, and Gateway, to name three – and these questions will frequently arise. Luckily, some basic economics provides some guidance.

The Economy, Pipelines, and “Value Added Jobs”

The first reason President Obama gave for rejecting Keystone was it “would not make a meaningful long-term contribution to our economy.” Of course, he’s right. No single project could cross that hurdle. For perspective, consider a recent research paper in the Quarterly Journal of Economics – a top journal – which finds the entire U.S. interstate highway system raises U.S. real GDP by about 1.1 per cent to 1.4 per cent (ungated version here). That’s the contribution of 76,000 km of highways, with annual building and maintenance costs of perhaps $100 billion. The contribution of a single pipeline that is orders of magnitude smaller in scale would essentially be a rounding error. (For a specific number on this, the pipeline may have added 0.02 per cent to U.S. GDP; based on this.) This isn’t an argument against the pipeline, but it is important to keep in mind that for any single project to make a “meaningful long-term contribution” to an entire economy is an unrealistic bar to set.

The President also cited the lack of long-term jobs associated with the pipeline. To an economist, the economic contribution of the pipeline (and indeed of any project) is not the number of workers it requires, but the service it provides: in this case, transporting oil from point A to point B. What matters is whether the value of the oil at point B is sufficiently greater than the value of the oil at point A – sufficient, that is, to cover the costs of operating the pipeline. Done efficiently, there will naturally be few jobs associated with it. If we wanted to maximize the number of jobs required to transport oil then, as University of Alberta economist Andrew Leach points out, we could establish a continental bucket brigade. Hardly sensible policy. Overall, economists do not typically think about the number of jobs, but instead think about the type of jobs. Policies will normally just shift employment from one sector or region to another, rather than sustainably increasing the total number.

Are some types of jobs better than others? During pipeline debates in Canada, one often hears references to “value-added jobs.” Consider the federal NDP position that Keystone will “ship thousands of jobs away” and that “exporting unrefined heavy oil creates no value-added jobs” (from here). No doubt the cancellation of Keystone will lead some to increase the call for domestic refining and upgrading. And it’s not just the NDP; parties across the political spectrum seem put a premium on domestic processing activities over resource extraction and export. Consider the slogan “BC Logs for BC Jobs,” or the billions in subsidies provided by the former Alberta PC government to the Northwest Upgrader. Of course, politicians are just responding to incentives. Polls consistently show broad public support for government support of domestic processing and for finding ways to “add value.”

But, what is a value-added job? A good rule of thumb is that value-added is basically just income. An activity that generates $1 billion in income for the workers and business owners (net of depreciation) is said to have a value-added of $1 billion. A “value-added job” is therefore a job that generates income. To say that something doesn’t add value is to say that it doesn’t generate income. So a job in raw extraction, which pays very well, is as much a value added job as any other.

You don’t need to take my word for it. Value-added per job is something we can measure. The following displays the five highest and lowest sectors by value-added per job, based on Statistics Canada data (source: here).


Oil and gas extraction has (by far) more value-added per job than most other sectors. It therefore can’t be said that a pipeline exporting raw oil comes at the expense of value-added jobs elsewhere. Of course, this does not suggest that extraction jobs are somehow “better” for the economy than other jobs. It equally does not suggest that jobs in the Hotel and Restaurant sector – which, as one can see in the figure, has a very small average value-added per job – are “bad”. The point I wish to make is that knowing how much value each sector creates per job is almost completely useless information for policy debates; perhaps even worse than useless, it’s distracting and misleading. (For a more detailed discussion, see here.)

A Distraction from Good Environmental Policy

The other aspect of the Keystone debate is, of course, the environmental implications of shipping oil extracted from Alberta’s oil sands. Many view opposing Keystone as synonymous with combating climate change. Some basic economics suggests there are far better approaches on offer.

Consider putting a price on carbon. British Columbia, for example, is a global leader with perhaps the best example of a broad-based and uniform carbon tax. B.C. charges $30 per tonne and covers approximately 70 per cent of all sources of emissions. Piecemeal regulatory approaches or subsidies for “green jobs” or “green infrastructure” are often costlier ways of achieving environmental objectives than simply putting a uniform price on carbon. For climate change, a tonne is a tonne is a tonne – and we should find the lowest cost sources of abatement. This is basic economics. (For more, see research material from Canada’s EcoFiscal Commission; in particular, this report.) As an added bonus, the revenue from a carbon tax can be used to lower other (worse) forms of taxation, such as personal and corporate income taxes. This was also B.C.’s approach. It doesn’t increase the size of government, or increase the tax burden on the average person, it just changes the composition of taxes we pay.

What does all this have to do with Keystone? For starters, the debate around Keystone distracted us from a sensible discussion about optimal environmental policies. There’s an opportunity cost of lobbying efforts, and we must use our political capital wisely. Time, energy, and money spent lobbying for one policy comes at the direct expense of another. A focus on blocking a single pipeline may actually hurt the environment by distracting attention from more effective and efficient policies. To those who want to see action on greenhouse-gas emissions with the smallest possible effect on the economy: don’t waste time debating one project or another, one sector or another, or one arbitrary emissions target or another; instead, push for a broad and uniform price on carbon and be done with it.

Support for this approach is growing, both here in Canada and around the world. Overall, about 12 per cent of global emissions are currently covered by some form of carbon price. The overwhelming majority of economists view this approach as the best route forward. (For polls of top economists on this point, see here or here.) The policy also makes for easy comparison with others. Instead of comparing emissions targets that may not ever be met, or arguing over this pipeline or that, we can compare prices. The higher the price and the broader the coverage, the more one is doing to combat climate change. Pipelines need not factor into the discussions at all.

The stakes are particularly high for Alberta, though so are the potential benefits. With a broad-based price on carbon, Alberta could confidently declare itself to be among the world’s leaders. Future pipeline projects may then no longer become lightning rods, but could be calmly and rationally evaluated on their own individual merits. Perhaps this is overly optimistic, but with the Paris talks only a few weeks away and with the Alberta government soon announcing its policy intentions, we won’t have to wait long to find out.


Trevor Tombe is an assistant professor in the department of economics at the University of Calgary. Follow him on Twitter: @trevortombe