I have a friend at the Census who is, in principle, allowed to use R or Python, but is definitely aware of a strong historic bias for commercial statistical sof

I have a friend at the Census who is, in principle, allowed to use R or Python, but is definitely aware of a strong historic bias for commercial statistical software--SAS mainly. That's the weak form of pro-commercial government bias. The strong form is that there are, I'm told, agencies in which only "certified" statistical software can be used.