I find this question to be a classic ridiculous slippery slope. For example: Do you count the foot print of the plane that takes employees on vacation bought using the company's salary? Or the foot print of the extra child they decided to bear thanks to the finical stability the company offers?
I feel these type of claims should not be taken literally but rather as "We made a substantial dent in our footprint".
I would expect companies to count plane emissions when their employees go on business trips, but not on holidays.
I would also expect companies to count emissions produced by their employees when they work remotely, which should be a part of the total emissions of their home.
Sure, they may not be able to get an exact number, and it will be lots of estimates, but it does not mean that we should accept some approximate statement such as "We're trying our best"
Should the company also reduce their carbon footprint calculations since paying employees more will make then more likely to own a tesla or other electric?
A given employee might or might not fly or whatever. But statistically, the average is going to be pretty predictable, and it's not intellectually honest to pretend that because one doesn't know individual behavior, the average could be anything.
https://www.epa.gov/greeningepa/greenhouse-gases-epa
Though I think the MS article is easier to understand. If you think this is not a reasonable or accurate way to measure emissions, there are some great peer reviewed journals you could write to.