Canada Markets

Challenges Lie Ahead for Statistics Canada's Production Estimates

Cliff Jamieson
By  Cliff Jamieson , Canadian Grains Analyst
Connect with Cliff:
Canadian crops tend to grow larger, as seen in this comparison of the percent change from the historical September survey-based estimates and the final November estimates released in December, although some crops tend to grow more than others. (DTN graphic by Nick Scalise)

With Statistics Canada set to release its final 2016 estimates for Canadian crop production on Dec. 6, the agency faces a challenge, given the challenging weather conditions which hampered harvest during the survey period which ranged from Oct. 21 through Nov. 13. In the earlier days of the survey, Saskatchewan producers were still dealing with the final 20% of the province's harvest, while Alberta farmers faced their final 25%.

"This is a very unusual situation. The way our surveys are designed right now, it's making it very hard to adapt the actual collection period in the risk of not making our deadlines," Yves Gilbert of Statistics Canada told

For many crops, final November estimates are generally larger than the historical survey-based estimates produced in September and released in early October. As indicated on the attached graphic, the red bars for the selected crops represent the five-year average percent change from the September survey-based estimates to the final November estimates released in early December. Over the 2011-through-2015 period, this increase was as little as 2.4% on average for barley, while as high as 14% for canola. Very seldom over this period has the November estimates for the major crops been reported lower than the previous September estimate.

A small part of this is a result of past September surveys excluding smaller-producing provinces, with past Statistics Canada reports indicating that production data from Newfoundland, Labrador, Prince Edward Island, Nova Scotia, New Brunswick and British Columbia are not included until the final estimates are released, accounting for about 2 to 4% of production, according to Statistics Canada estimates.

2016 will perhaps result in new challenges, given the elimination of the September survey which was entirely replaced this year by the model-based analysis which was released Sept. 20. This will be the second year of using this methodology, which was run parallel to the survey method in 2015. Time will tell whether this will smooth the results of the successive forecasts over time, although 2015 model-based data shows that the original forecasts have almost all been revised since their initial release, while still point to variability when compared to final data released in December.


DTN 360 Poll

This week's poll asks what you think we can look forward to in row-crop prices (soybeans and corn) over the balance of the crop year. We encourage you to share your thoughts on this poll, found at the lower-right of your DTN Home Page.

Cliff Jamieson can be reached at

Follow Cliff Jamieson on Twitter @CliffJamieson



To comment, please Log In or Join our Community .