top of page

Pipelines create supply. Policy shapes the emissions

  • Writer: Dave Sawyer
    Dave Sawyer
  • 1 hour ago
  • 3 min read

The Canada–Alberta MOU brings into focus an old debate: do pipelines simply move oil, or do they incent new production and emissions?  The data indicates pipelines unlock supply, which makes emissions policy matter.


Draft working paper: January 11, 2026


Do pipelines create new oil production and emissions, or do they just move existing barrels around at lower cost? The Canada–Alberta MOU puts this question into sharp focus.


To answer this, we need to look at how Alberta's production changed over time when new export capacity came online, compared to what would have happened without that capacity.


The basis of the wonkery

Econometric analysis is well-suited to address this question .To isolate the causal impact of pipeline capacity on production, I adopt a difference-in-differences approach that compares changes in Alberta’s production to changes in Saskatchewan’s production before and after pipeline expansions. This approach controls for persistent differences between the two provinces, accounting for factors that do not change over time, including geography and infrastructure, as well as time effects that impact both provinces, including oil prices, COVID disruptions, and the post-pandemic recovery. Once these fixed effects are accounted for, any remaining differences reflect the incremental impact that pipeline capacity likely induced.


Given the importance of the WCS–WTI price gap, and the circular relationship between capacity, prices, and production, the approach isolates the causal effect of pipeline capacity working through prices.

Data for the analysis include national crude-by-rail shipments, oil production, the WCS–WTI price gap, and the timing of the Line 3 replacement (L3R) and TMX coming online.


The work estimates the production and emissions impacts of recent major pipeline expansions using public data and standard quasi-experimental methods; while it has limitations, the direction and order of magnitude of the results are robust.


Incremental barrels flowed, rail fell away

The evidence is compelling. After L3R entered service in October 2021, Alberta’s production grew incrementally by about 343,000 barrels per day more than Saskatchewan’s (p < 0.0001), even after controlling for fixed effects in both provinces. When TMX came online in May 2024, Alberta’s production grew relative to Saskatchewan’s by a further 201,000 barrels per day (p < 0.005). Taken together, pipeline expansions are associated with roughly 544,000 barrels per day of incremental Alberta production that cannot be explained by market recovery or long-run provincial differences.


Pipelines also changed how oil moved. National rail exports declined by roughly 129,000 barrels per day as new pipeline capacity enabled producers to shift to lower-cost pipelines. But this rail displacement is far smaller than the increase in production observed in Alberta. The implication is straightforward: pipelines did not just substitute for rail; they relaxed a binding constraint on total output. We know this from Alberta’s earlier use of production curtailment to stabilize the WCS–WTI price spread in early 2019.


Emissions followed

We then translate this incremental production into emissions. If emissions per barrel were held constant, the additional volumes implied by pipelines would add roughly 13 megatonnes of upstream emissions per year. But emissions intensity has been falling over time as technology improves and abatement is driven by carbon pricing, methane regulations, and other policy supports. Since 2019, upstream emissions per barrel in Alberta have declined by roughly 1 to 1.5 percent per year, reflecting efficiency gains. Accounting for this decline reduces the net upstream emissions increase to about 8.1 megatonnes per year, implying a technology and policy dividend of roughly 4.9 megatonnes.


Emissions policy matters

Pipelines clearly unlocked volumes by removing a binding infrastructure constraint. We see this directly in the data. But we also see a rise in emissions over this period. The key takeaway is that while more pipelines lead to more upstream production, the emissions outcome is not certain. Whether rising production comes with rising emissions depends on the strength of climate policy and the pace of technological improvement. Pipelines move oil. Policy is a key determinant of the emissions outcome.



Download here.

Github repository with data, model, and user guide available: https://github.com/EnvirEco/pipelines/tree/main

Comments on draft welcome.

 
 
 
Featured Posts
Recent Posts
  • Twitter Basic Square
  • Twitter Basic Black

© EnviroEcononomics 

bottom of page