StatsCan Needs To Get It Right – for Sep. 9, 2010

It’s rare to find a grain farmer with anything good to say about Statistics Canada. At best, the field crop production reports are viewed as a waste of time and money. At worst, the reports are viewed as a government plot to manipulate grain prices.

For each of its reports, StatsCan surveys a pile of producers, but all too often the results are disputable. Take the latest report, the July 31 estimate of production for the principal field crops. In Saskatchewan alone, a total of 4,838 farmers were surveyed from July 26 to Aug. 3.

No farmer relishes a call from StatsCan. They make participation sound mandatory, although I’ve never heard of any ramifications for growers who refuse to participate. It doesn’t help that the poor folks charged with doing the surveys typically couldn’t tell you the difference between wheat and canola.

Some growers pull numbers out of the air. Others tell blatant lies and brag about it later. From this comes a report that can move grain prices either up or down.

The process is flawed. Even for producers willing to participate, it isn’t realistic to think every grower will have all his or her acreage and expected yield numbers at the tip of the tongue whenever the phone happens to ring.

Not surprisingly, the survey results are increasingly being questioned. The July 31 production estimate (released Aug. 20) has summerfallow (unseeded) acres at 9.67 million acres in Saskatchewan. In 2009, summerfallow acreage was at 4.1 million.

The provincial government has estimated unseeded acreage at eight million. According to StatsCan, the increase in unseeded acreage is less than 5.6 million. That’s a big difference. Who’s right?

A lot of land that was seeded was subsequently flooded out, but this doesn’t seem to be reflected in the StatsCan numbers. Their estimates of seeded acres versus harvested acres show about the normal variance. Where did all the flooded acres go?

With more acres in production than expected, production estimates are also higher than expected on many crops. That has a depressing effect on prices – one of the main complaints lodged by producers. It seems StatsCan reports are always bad news for prices.


Another problem is timeliness. StatsCan is surveying farmers three weeks before the actual report is released. That’s an eternity in the growing season. The world has moved to instant communications, but StatsCan is still stuck in the days of the Pony Express.

A lot of observers argue that the federal government should have no role to play in publishing crop production estimates. While it’s easy to buy into that sentiment, consider what would happen if that came to pass.

Right now, everyone examines the StatsCan numbers and either agrees or disagrees, but the report becomes the basis for discussion and analysis. Amazingly, StatsCan often turns out to be more accurate that we give them credit for.

Without an official estimate, you can bet all sorts of surveys would still be done by agribusiness, but this information wouldn’t necessarily be public. Under the current system, farmers have just as much access to the information as anyone else. It’s a level playing field.

Like other civilized countries around the world, Canada should continue to issue crop acreage, yield and production reports, but confidence must be restored in the integrity of the process and the numbers. What we have now is becoming a colossal waste of resources that’s doing as much harm as good.

– Kevin Hursh is a consulting agrologist and farmer based in

Saskatoon. He can be reached at [email protected]

About the author



Stories from our other publications