The Trivers–Willard theory postulates that if a male in good condition at the end of the period of parental investment reproduces more successfully than a sister in similar condition, the mother should overproduce sons. Although this theory has been tested empirically in a wide variety of species, often producing equivocal results, few studies have analyzed whether the study populations conform to its assumptions. Here, we analyzed sex differences in lifetime reproductive success in a wild population of a polygynous bird, the Spotless Starling (Sturnus unicolor), and tested the Trivers–Willard prediction. We found that lifetime reproductive success was influenced by the body mass at fledging in males but not in females. However, although mothers with higher body mass produced heavier fledglings, they did not increase the proportion of sons. By contrast, we found that mothers with higher body mass produced heavier male fledglings and, after fledging, sons increased their body mass more than daughters. Our results suggest that there are sex-specific selective pressures acting on fledgling body mass. This may influence the sex-allocation tactic of the mother (i.e. biasing parental investment toward sons) or the rate of growth and the intensity of sibling competition (i.e. favoring sons to grow faster and hoard parental resources). Therefore, increasing the proportion of sons may increase the cost of parental investment and sibling competition, which may instead favor even sex ratios. Our study reinforces the idea that it is necessary to take into account the complexity of vertebrate life histories to make further predictions on sex-allocation strategies.