I'm sure that makes a small difference, but I doubt it's enough to meaningfully change the conclusion. 1. The BLS knows this, so they get their data from multi

I'm sure that makes a small difference, but I doubt it's enough to meaningfully change the conclusion. 1. The BLS knows this, so they get their data from multiple services--not just government agencies. 2. The IRS knows this, so are more likely to audit people who work in industries that receive tip income. 3. If the BLS numbers were vastly underestimating bartender income because of unreported cash tips, you would expect bartender income to have risen much faster than other similar low wage jobs over the last 20 years as cash has become less common. This hasn't happened.