Assume you are trying to study the effect of democracy on trade. You believe that it does this by decreasing tariff rates. I.e. democracy causes reduced tariff rates, which increases trade. Do you need to include tariff rates as a control variable? Group of answer choices It might help to get more precise estimates, but including it is not required No, you definitely should not include it You should not include it initially, but running a second model with it included might help identify mechanisms Yes, you definitely should include it