A key tool used to assess reproductive trade-offs in birds is brood size manipulation (BSM) experiments. Most BSM studies have examined the influence on short-term measures of reproductive output. Seldom evaluated are the effects on long-term fitness proxies under temporally or spatially varying environments. Unpredictable environments may affect reproductive trade-offs by altering the value of the brood or hampering optimization of reproductive effort. We reduced or enlarged broods of 140 male Tengmalm's owls Aegolius funereus by one chick during their first lifetime reproductive event. Males differed in age and bred in environments that varied in quality spatially (habitat structure) and temporally (abundance of main food). We measured the short-term (nestling number and condition) and long-term fitness proxies (survival, lifetime fledgling and recruits produced) until all experimental males disappeared from the population. BSMs did not affect fledgling number or condition, but in enlarged broods, offspring condition was lower in territories with a high proportion of agricultural fields. Importantly, no obvious impacts on long-term fitness proxies emerged; lifetime fledgling and recruit production of males did not differ between the BSM treatments. Thus, the primary caregiver (i.e. Tengmalm's owl males) passed increased reproductive costs to their offspring, which is in agreement with other studies investigating intergenerational reproductive trade-offs in species of intermediate lifespan. Reluctance to accept increased current reproductive costs in these systems highlights the potential for sexual conflict in bi-parental care systems in which one of the pair is the primary caregiver.