Is the cost of reproduction different between males and females? On the one hand, males typically compete intensely for mates, thus sexual selection theory predicts higher cost of reproduction for males in species with intense male-male competition. On the other hand, care provisioning such as incubating the eggs and raising young may also be costly, thus parental care theory predicts higher mortality for the care-giving sex, which is often the female. We tested both hypotheses of reproductive costs using phylogenetic comparative analyses of sex-specific adult mortality rates of 194 bird species across 41 families. First, we show that evolutionary increases in male-male competition were associated with male-biased mortalities. This relationship is consistent between two measures of mating competition: social mating system and testis size. Second, as predicted by the parental cost hypothesis, females have significantly higher adult mortalities (mean ± SE, 0.364 ± 0.01) than males (0.328 ± 0.01). However, the mortality cost of parental care was only detectable in males, when the influence of mating competition was statistically controlled. Taken together, our results challenge the traditional explanation of female-biased avian mortalities, because evolutionary changes in female care were unrelated to changes in mortality bias. The interspecific variation in avian mortality bias, as we show here, is driven by males, specifically via the costs of both mating competition and parental care. We also discuss alternative hypotheses for why most birds exhibit female-biased mortalities, whereas in mammals male-biased mortalities predominate.