Cost of display has emerged as a prominent theme in some recent discussions of models that attempt to explain the evolution of exaggerated sexual display traits in males. Here I use comparative and experimental information from an 8-yr field study of the satin bowerbird (Ptilonorhynchus violaceus) to test predictions from sexual selection models that focus on the cost of male display. Two of these models predict a high absolute cost of male display when extended to the satin bowerbird. Eight different tests showed no evidence of the predicted high cost of display, and four produced significant results in the opposite direction. These results are consistent with the claim that truthful advertising need not be costly and Gilliard's suggestion that there has been selection for reduced cost of male display among bowerbirds. Evidence showing that exaggerated male display traits do not have a high absolute cost may help resolve the long-standing question of how these displays evolve.