I agree with many comments here, but one of the terms seems absurd: > Not use models that determine geographic effects narrower than at a state level They are

I agree with many comments here, but one of the terms seems absurd: > Not use models that determine geographic effects narrower than at a state level They are forbidden from having different models for different cities?? This seems extreme to me. Any background on this?