We test whether labor supply responds symmetrically to wage increases and decreases using a randomized real effort online experiment. The results show that wage increases have smaller effects on labor supply than wage decreases of equal magnitude, especially on the extensive margin where the response to a wage decrease is twice that to a wage increase. This finding suggests that labor-supply responses to wage changes are asymmetric. We discuss the potential mechanisms behind our results including standard models of labor supply, reference dependence in consumption and reciprocity