High overall employment numbers

My apologies if this is a well-trodden topic (I searched around a bit for an existing thread and didn't see one), but could you comment on the overall US employment number in 2010 in IMPLAN being 173 million? This is considerably higher than most official series for employment (the FRED series CE16OV, for example, was 139 million in 2010). These series typically exclude some groups (e.g. the military), but that's not enough to account for the difference. What is the source of this large difference? Does this suggest that, especially for smaller regions, one should compare overall job effects coming out of IMPLAN to employment numbers within IMPLAN rather than other official sources (e.g. the jobs impact of this project is equivalent to 1.1% of existing employment in county X)?