I believe he's making two claims: 1. The magnitude of the Canada -> US migration flow has lowered 2. The magnitude of the US -> Canada migration flow has r

I believe he's making two claims: 1. The magnitude of the Canada -> US migration flow has lowered 2. The magnitude of the US -> Canada migration flow has raised. I am unsure if either claim is true, but both claims can be true without the net migration flow switching signs.