Predicting the embodied scope 3 carbon dioxide equivalent (CO2e) emissions from purchased electricity for end users in the United States is challenging due to electricity transmission within interconnected power grids. Existing methods only focus on large aggregation areas, thereby ignoring potentially significant emission factor (EF) variations, so this study proposes a novel method to translate the CO2e emissions from the balancing authority (BA)-level to the county-level by utilizing explicit finite-difference theory for electricity flow predictions, and then employing economic input–output theory to evaluate the scope 3 embodied lifecycle CO2e emissions. Results show that the generation-based EFs at the BA-level range from 0.007 to 0.905 MT-CO2e/MWh with a mean value of 0.400 MT-CO2e/MWh and a standard deviation of 0.229 MT-CO2e/MWh. The consumption-based EFs at the BA-level range from 0.008 to 0.836 MT-CO2e/MWh with a mean value of 0.378 MT-CO2e/MWh and a standard deviation of 0.019 MT-CO2e/MWh. Results also show that sixteen BA consumption-based EFs deviate by more than 20% compared to their generation-based EFs, which indicates the significance of accounting for electricity interchanges in emissions quantification processes. A larger range of possible consumption-based EFs is revealed at the county-level: 0.007 to 0.902 MT-CO2e/MWh, with a mean value of 0.452 MT-CO2e/MWh and a standard deviation of 0.123 MT-CO2e/MWh. Results also indicate significant variations in EFs of counties within each BA: 20 BAs have county-level EFs range greater than 0.1 MT-CO2e/MWh, 13 BAs have county-level EFs range greater than 0.2 MT-CO2e/MWh and 6 BAs have county-level EFs range beyond 0.3 MT-CO2e/MWh.