We examine the evolutionary stability of year‐round residency in territorial populations, where breeding sites are a limiting resource. The model links individual life histories to the population‐wide competition for territories and includes spatial variation in habitat quality as well as a potential parent‐offspring conflict over territory ownership. The general form of the model makes it applicable to the evolution of dispersal, migration, partial migration, and delayed dispersal (offspring retention). We show that migration can be evolutionarily stable only if year‐round residency in a given area would produce a sink population, where mortality exceeds reproduction. If this applies to a fraction of the breeding habitat only, partial migration is expected to evolve. In the context of delayed dispersal, habitat saturation has been argued to form an ecological constraint on independent breeding, which favors offspring retention and cooperative breeding. We show that habitat saturation must be considered as a dynamic outcome of birth, death, and dispersal rates in the population, rather than an externally determined constraint. Although delayed dispersal often associates with intense competition for territories, life‐history traits have direct effects on stable dispersal strategies, which can often override the effect of habitat saturation. As an example, high survival of floaters selects against delayed dispersal, even though it increases the number of competitors for each breeding vacancy (the “habitat saturation factor”). High survival of territory owners, by contrast, generally favors natal philopatry. We also conclude that spatial variation in habitat quality only rarely selects for delayed dispersal. Within a population, however, offspring retention is more likely in high‐quality territories.