The Bureau of Labor Statistics (BLS) uses two distinct surveys to measure the number of jobs in America, a payroll survey that measures the number of people employers have on their payrolls and a household survey that measures the number of individuals who report being employed. Though analysts focus on the payroll estimates, the household survey has recently been painting a surprisingly different picture of the U.S. labor market. The often-cited payroll survey indicates that the number of jobs has declined by 1.0 million since the end of the recession in November 2001, while the household survey indicates that the number of employed people has increased by 1.4 million. Economists cannot yet fully explain this 2.4 million “jobs gap,” but small businesses and, in particular, self-employment appear to be significant factors.