This paper empirically examines how consumers react when a company marks a product with a gay label. The company under scrutiny is one of the largest video game developers in the world, and the labeled product is a popular video game character. We use a regression discontinuity design to exploit the quasi-experimental setting. The main finding was significant drop in demand for this character and a return to previous levels after approximately 3 months. Possible mechanisms and dynamics were explored by (...) analyzing demand for other characters and by surveying consumers. The results are consistent after performing robustness checks based on grid search, subsampling, and placebo tests. (shrink)