Skills, Migration, and Urban Amenities over the Life Cycle

We explore how much sorting across cities based on worker skill is driven by a desire to live in high-amenity cities, and if so, whether this sorting has a life-cycle component. We use data from the restricted-access geocode files for three cohorts of the National Longitudinal Surveys of Youth (NLSY), supplemented with data on time use and household expenditures. We estimate a quality of life index, which captures the degree of local amenities and labor productivity within a metropolitan area, using the methodology from Albouy (2012, 2016). We find that, like consumption expenditures, the average quality of life of an individual’s metropolitan area is hump-shaped over the life cycle, though it peaks much earlier. The observed life-cycle changes in average quality of life reflect migration decisions. Variations in average quality of life are driven primarily by changes in local amenities (and are reflected in changing housing prices across metropolitan areas) and are largest for the high skilled. We argue that the early hump-shape in quality of life reflects a bunch of amenity demand towards the earlier stages of the life cycle, and that this bunching is driven by expectations of child rearing, which diminishes the household’s demand for local amenities. We present evidence on time use and expenditures over the life cycle to support this argument. Time and income spent on local amenities is U-shaped over the life cycle (i.e., highest prior to having children and during retirement), and they are highest for the high skilled. In contrast, time and income spent on leisure activities at home are constant over the life cycle, and are highest for the low skilled. Thus, migration decisions have both a sorting component, which is driven in part by the desire to live in high-amenity cities, and a life-cycle component, which is driven by the presence of household children.