After the federal approval of Kinder Morgan's controversial Trans Mountain Pipeline Expansion (TMX), Alberta Premier Rachel Notley came to B.C. to sell the pipeline's economic benefits. She claims B.C. will get a $1 billion per year boost in GDP as a result of the pipeline, as well as thousands of jobs in both construction and operations once the pipeline is built.
There is good reason to believe, however, that the incremental income and employment gains for ordinary working British Columbians are likely to be substantially less. The project would lead to a small increase in total GDP in the province, but the main benefits are in the construction phase, and even there the benefits have been grossly inflated. In the long run, we can only bank on about 50 new permanent jobs.
The economic case for the pipeline was commissioned by Kinder Morgan, and published in three studies by the Conference Board of Canada, all using a flawed approach called input-output modelling. This approach is routinely used by governments and industry to justify large projects, and is biased towards producing exaggerated numbers.
The input-output approach is not the same as cost-benefit analysis. It does not consider economic, social or environmental costs; it merely counts all economic activity as a benefit. For example, many are concerned about the catastrophic damages that could occur in the event of a tanker accident or spill on land. The input-output approach ignores these risks, which carry potential costs in the billions, but counts increased jobs and income as benefits.
While spills are of a probabilistic nature, the pipeline will also facilitate increased greenhouse gas emissions in Canada (from extraction and processing) and in overseas markets (from combustion for energy). Based on estimates of the damages from GHG emissions, the increased carbon in the atmosphere amounts to at least $4 billion per year in damages.
Similarly, a more balanced assessment would look at any adverse impacts on existing economic activity -- such as the impact of tanker traffic on fishing or tourism. Some damages, like the potential loss of B.C.'s south coast whale populations, cannot be put into dollars except in crass human terms (the loss of a whale-watching industry).
A 2014 cost-benefit analysis from SFU's School of Public Policy found that in the event of a major spill on land or at sea "the costs will exceed, or greatly exceed, the benefits for B.C. and Metro Vancouver." The authors conclude that "the lion's share of the benefits flows to KM/TMP, the Alberta tar sands producers and Alberta, whereas the citizens of B.C., and Metro Vancouver in particular, will bear the lion's share of the risks and receive very small benefits."
Even in its limited "benefits only" approach, there are a number of shortcomings of input-output modelling. This approach should be interpreted with caution as it represents a simplified extrapolation of existing economic relationships. The model assumes no economies of scale, and no changes in technology or prices. Nor does it consider the economic impacts of alternative investments in, say, green infrastructure. (See my cost-benefit analysis of the Enbridge Northern Gateway pipeline for a more details on input-output methodology.)
In the Conference Board reports, there is little transparency. Only the modelling results are reported, and it is hard to square the large numbers they produce with even Kinder-Morgan's own estimates.
Most of the economic benefits from the pipeline would come during the construction stage. Buried in Kinder-Morgan's National Energy Board submission is its estimate of 4,475 workers at peak construction, although this is for one month only. Averaged over a two-year construction period, about 2,500 full-time workers will be employed (volume 5B, page 2-16). In contrast, the Conference Board model estimates peak construction activity at 12-13,500 jobs for the two years of intensive construction -- five times Kinder Morgan's own estimate.
In addition to direct employment in pipeline construction, there are potential indirect (or upstream) jobs in manufacturing the steel and pipe, transporting it into place, as well as other associated professional work. If the pipe was manufactured in Regina, some 1,000 jobs could be supported for two to three years. But there is no guarantee this would be the case as pipe could also be imported from Korea or elsewhere.
Importantly, input-output models assume that workers getting the jobs would be otherwise unemployed. In reality, the vast majority of those hired would simply be working elsewhere in the absence of the pipeline project. We should also not jump to the conclusion that workers employed in B.C. would come from B.C. In the pipeline construction industry, there are crews of specialized workers who move around to where the work is. It is possible that temporary workers from outside B.C. would fill positions.
For ongoing pipeline operations, Kinder Morgan's application notes that the existing Trans Mountain pipeline supports 100 jobs in Alberta and 100 jobs in B.C. Kinder Morgan projects an additional 90 workers as a result of the expansion, 50 of which would be in B.C. Yet, the Conference Board estimates new employment for the pipeline expansion at a much higher 443 jobs (313 in B.C.), and then makes some heroic assumptions to inflate this to 7,600 jobs per year in B.C.
These results include some padding by including "induced employment," representing about one-quarter of the employment gain. This is supposed to represent income going to workers, which is spent in the local economy and therefore supports other employment. Estimating induced effects is problematic because it is based on a false assumption that the pipeline project would be hiring people who would be otherwise unemployed and spending nothing. It also assumes that worker expenditures in the local economy increase other employment in a linear fashion (e.g. doubling sales at the grocery store means a doubling of workers at the store).
The Conference Board also rolls in additional income to oil sands producers that would come from selling Alberta bitumen at allegedly higher world prices than garnered from the U.S. But as pointed out by economist Jeff Rubin, prices for similar heavy oil product in Asia are somewhat less than in North America, not more. It is true that Alberta's heavy crude oil from the tar sands sells at a lower price in the United States compared to other oil. But the lower price reflects selling an inferior product that requires more intensive refining, as well as higher transportation costs to get it to market (compared to conventional lighter oils).
This final bit of padding to get to the Conference Board's $1 billion in increased GDP per year assumes $50 billion in additional after-tax profits to the oil and gas industry over 20 years. It is assumed that $8.5 billion of that amount is paid out to Canadian shareholders as dividends that then get spent in the economy (plus $6 billion paid out to foreign shareholders). The Conference Board also assumes that $35 billion in retained earnings leverages an additional $22 billion in borrowed funds, and that this full $57 billion in invested in Canada's oil and gas sector. Both the dividends and the new investment are run through the input-output model again -- yielding 40 per cent of the alleged gain in B.C.'s GDP and 60 per cent of the total employment gain.
Simply put, the assumptions made at every stage of this calculation are dubious, leading to exaggerated numbers that are simply too good to be true. In its totality, this modelling exercise makes a mockery of economic benefits, neglects real and potential costs, and is being used to misinform British Columbians about the impact of the TMX pipeline.
Marc Lee is a co-investigator with the Corporate Mapping Project (CMP), a research and public engagement initiative investigating the power of the fossil fuel industry. The CMP is jointly led by the University of Victoria, the Canadian Centre for Policy Alternatives (B.C. and Saskatchewan Offices) and the Parkland Institute. This research is supported by the Social Science and Humanities Research Council of Canada (SSHRC).
This piece originally appeared on the B.C. CCPA's Policy Note blog.