I've read a bit more about the background to this now, and they hired someone in December who immediately went on an internal advocacy campaign for their personal politics. So, all of this, is the cost of hiring a "wrong" person.
Having to replace (currently) 30%+ of your org increases that risk, in addition to the initial costs of hiring all those people and getting them up to speed.
Given that they seem to view politic discussions and committees as non-productive and that most of the people that left did so because these were important to them, they'll probably account for parts of their work and their leaving as "lack of productivity" and "hiring the wrong people" already.
EDIT: From a company perspective - not judging either way.
All the people that left that I knew were incredibly talented and productive people, they are going to be a nightmare to replace.
> it's 50% better for the 2/3 of remaining employees.
Only in a very simplistic view.
Basecamp runs very lean, ~58 people before this incident, which for the number of customers (large) is a small amount.
Basecamp is going to need to replace ~100% of those people. That means hiring costs, costs for hiring the wrong people, and lack of productivity.
This will most likely cost them more money than they "save".