Many couples want to retire together even if spouses differ in age. Drawing on theories of leisure complementarity, gender roles and social status, this article uses comprehensive Swedish register data from 2002 to 2010 to explore synchronised retirement and its association with spousal age differences and other socio-demographic factors. Synchronisation rates in dual-earner couples (N = 83,986) were 10 per cent for retirement the same calendar year and 25 per cent for retirement the same or the following year. Contrary to theoretical expectations, synchronisation was more common in women-older couples than in men-older couples, although this was largely a consequence of the skewed distribution of age differences. Moreover, spouses' education, incomes, assets, employment and health were differently associated with synchronisation in same-age, men-older and women-older couples. In the total population, average retirement age differed very little between synchronising couples and other couples. Yet women who synchronised retired at an earlier age than other women, whereas men who synchronised retired later than other men. This was partly an effect of the predominance of men-older couples, but men in men-older couples were also more likely than women in women-older couples to delay retirement in order to synchronise.