In this paper, we scrutinise Oswald's evidence that home-ownership and unemployment are correlated across the US states. In order to abstract from state fixed-effects in levels, we analyse the cross-sectional variation in changes in home-ownership and unemployment rates between 1970 and 1990. After duplicating (nearly) Oswald's result, we illustrate the importance of weighting the state observations by the fraction of total US households in 1970 that resided in the respective states and of abstracting from the ageing of the population between 1970 and 1990: this causes Oswald's correlation to to disappear. Secondly, we estimate the relationship for six different age-classes and for household heads and total population. We find that the relationship is non-existent for both young households and old households, but exists for middle-aged households. Young households have accumulated little wealth and have had less time to become attached to the geographical area than middle-aged households and thus are more likely to respond to unemployment by relocating. Older households' employment cannot be greatly affected by home-ownership because their members are largely not in the labour force. Unemployment rates of household heads are affected less by tenure than those of the population as a whole.
