[wenc]

You limit imported new infections and can focus on slowing community spread. Due to the exponential growth rate, this absolutely makes a difference.

In the past few days, Singapore and other countries have announced they are also adopting this strategy. Singapore too has a complex highly-interdependent border situation with neighboring Malaysia and has to make an exception there as well.

[colechristensen]

My point being that infected already exist here broadly, presumably in every urban area. The small number of people crossing borders wouldn't seem to make any difference, the goal should be on the more local levels of transmissions (closing restaurants or slashing capacity, hygiene requirements, closing schools/workplaces/events, i.e. venues for transmission) risking adding a few infected individuals to a large population of infected individuals seems to be the wrong step, or much much too late.

[bawolff]

That's probably not true (yet). From what i have read, there are reports of community spread in bc, alberta, and ontario. That's a lot of Canada but certainly not all of it. If a significant portion of cases are still travel related, closing the border makes sense imo.

[bobthepanda]

If a country had already hit its peak, new infected travelers potentially infecting they came across (customs, transport, at their lodging) might cause new local flareups, no? At which point you might have to wait for a new peak to subside.

[eftychis]

Unfortunately (or fortunately because other phenomena are based on this also -- see gossiping protocols), that is true. Once you have a critical population, related with the connectivity of the subgraph, the propagation is self-sustained, that is everyone is going to get infected in a SIR model or similar. One could argue that you might change the curve slope a bit. And I guess that is the bet here.

[JabavuAdams]

Roughly 250k Canadians cross the Canada-US border per day. Not sure about US citizens.