Newborn animals do not have a fully functional immune system and are thus impaired in their ability to fight parasites. Mothers can therefore increase the survival probability of their young by providing them with passive immunity, e.g. in the form of maternal antibodies transferred via the placenta or the eggs. The maternal responses are only induced when parasites are present, and have been observed not only in vertebrates but also in invertebrates. However, while these parasite-induced maternal effects are known to reduce the harmful effects of common parasites, they may also impose costs for the young, either because the maternal response impairs parental performance, or because maternally transmitted products moderate offspring development. We experimentally tested these two hypotheses in a wild great tit population. We exposed birds to a common ectoparasite before egg-laying to induce the maternal response, and thereafter separated egg-mediated maternal effects from effects on post-laying parental performance by cross-fostering whole clutches. To assess the costs of this response without its confounding benefits, we kept nests free of parasites after hatching. Since the costs of maternal effects can be expressed differently under relaxed and harsh rearing conditions, we simultaneously manipulated clutch size. First, parasite-exposed parents raised lighter young, suggesting that parasite defence or the induced maternal response are costly to the parents and reduce their capacity to raise young. Second, under relaxed but not under harsh rearing conditions, young with the flea-induced maternal effect were heavier and were in better body condition than controls, suggesting that the maternally transferred products can be allocated to physiological functions beyond parasite defence.