Basically, grading is a human function. It's a visual function. You throw a sample on a desk and you have a set of parameters that the Grain Commission comes up with annually that you compare your grades to. Obviously in a year like this, when the quality is very good, there is very little in the way of parameters, because there's no variation. In a year when quality is bad, what marks a number two and marks a number three can be simply somebody's idea.
The differences often occur between Saskatoon, or the inland graders, who are probably more familiar with numbers and maybe a little more sympathetic, and they get used to looking at it.... What happens in most cases is that when that product gets to the coast, Vancouver grades it lower. It gets kicked back, and then the process is for rechecking to Winnipeg. At this point I don't know what the reference point is to Saskatoon, but I think the big difference is there.
As for a percentage of cases when it happens, I can't state one. Obviously in a poor-quality year it happens more often; in a high-quality year it happens less. There is now imaging technology that will take out the human error.
But I think the big point here is not how much it happens; it's how the problem is dealt with when it happens. The best thing I see in this committee is the ability—which has escaped us—to deal with that problem more effectively, in an expedient and quick manner.