It ranges from 14 to 22 weeks, depending on the regional rate of unemployment, which I assume is higher in your region. I have the data to show you.
In such cases, the earnings are divided by the larger number rather than the actual weeks worked, lowering benefits. The divisor rule can disadvantage temporary workers or people in an intermittent situation, so it's not surprising that many people are arguing for this best 12 weeks formula.
I have suggested that the best 12 weeks--I abbreviated my comments on that--should be over a longer period of time. Because often right before a layoff, or to avert a layoff, many employers will shorten hours. I would support a best 12 weeks scenario, where one can count the weeks in a longer period.
While I think the 14-week pilot was very instructive to us, as researchers, and political representatives, I would say that the best 12 weeks over a longer period would deal with the problem. That would mean that people would accept work for shorter hours if that was all that was on offer, knowing that it wouldn't disadvantage--