Why manipulating job statistics is unlikely and a bad lie for the president 🤥

The US Department of Labor compiles labor statistics by set formulas long set out by regulation and law. It uses data collected by all 50 states based on mandatory reporting of hiring and layoffs as part of the Unemployment Insurance program.

Every state collects their own data on unemployment based on those mandatory reports of an employer that offers unemployment insurance, which is most with a few narrow exceptions like entrepreneurs who own their own business and farms that employ their owners. That data has to be accurate, otherwise it’s insurance fraud.

It’s not to say data collection is perfect. Often it takes months to get all of the data collected and computed from all employers even in the age of computers. When there is not full data available, projections are made and those projections based on incomplete data. That can lead to revisions up and down as more data comes available.

The highly federated nature of the unemployment data collection makes it unlikely states could act in a conspiracy to distort the data, as such a conspiracy would quickly be leaked to the press. A lot of people in a room are bad at keeping secrets. Not to mention it’s hard to get everyone to agree on a certain way to rig things.

Casting doubt on the unemployment data really cuts both ways. If Trump appoints a loyalist to oversee labor statistics, people will be very skeptical if there is a good jobs report going forward, even if that’s that the data legitimately points to. Instead, many will claim the numbers are fake even if they are what the data shows.