Part-time work has vastly expanded in most OECD labor markets during the last decades. At the same time, full- and part-time wages have grown increasingly apart, leading to a substantial raw part-time wage penalty. Using quantile regression methods, this paper analyses the female part-time wage gap across the wage distribution and over time (1990-2009), while controlling for sample selection into full- and part-time employment. The estimated part-time gap is much lower than the raw gap. Nonetheless, a persistent part-time penalty is found for the lowest wage quartile. The wide divide between the observed and estimated part-time gap is due to strong positive selection into full-time work, which increases over time. On the contrary, sample selection into part-time employment goes from being positive at the beginning of the 1990s to disappearing by the end of the 2000s, even turning negative at the lower end of the distribution. An exploration of potential mechanisms reveals a large prevalence of job mismatch at the lower end of the part-time wage distribution as well as rising differences in the distribution of job tasks between full- and part-time employment.