This paper examines whether labor market frictions can explain the level and growth of the college wage premium in the US. I develop a novel model where both capital skill complementarity and differences in the search frictions faced by college and non-college workers drive the college wage premium. The presence of search frictions, and hence monopsonistic power, provides a range of explanations for rising college premiums not present in competitive models i.e. changes to relative job offer rates, to firm heterogeneity or to bargaining power between education groups. College workers enjoy substantially lower job destruction rates and higher job offer rates than non-college workers, which generates the presence of a significant, and relatively stable, college wage premium in my model. I also find that bargaining strength, as captured by unionization rates, starts off at similar levels for college and non-college workers but declines more severely for non-college workers. This trend explains a substantial portion of the growth in the college wage premium in my baseline model.