Mobility Measures

Geographic mobility is a celebrated feature of American life. Deciding where to live is seen not only as a key personal freedom, but also a means of economic advancement. Millions of Americans move each year over great distances. But while this right to travel is safeguarded by the Constitution, these mobility decisions are not entirely free. In terms of moving long distances, employment and family reasons are central, and a regime of employment and family law “mobility measures” play a significant role in regulating why and how we move. This Article first sets forth this new framework of “mobility measures,” which are constituted by employment law sorting (moving across employers and space for employment purposes) and family law clustering (moving with a legally defined portable family unit). These mobility measures not only enable and facilitate long-distance moves, including with subsidies to the tune of billions of dollars a year, but they motivate these moves to take a particular form: to move for employment purposes, only taking our nuclear family with us. In this way, we are encouraged by the law to move, yet the law prevents us from mitigating the disruption caused by the move. So while mobility has its benefits, this Article argues that it has underappreciated costs. Long-distance moves destroy place-specific investments with our closest supporters that are crucial for everyday functions, as well as economic productivity. These relationship and economic costs harm all long-distance movers, but weigh particularly heavily on one group — women. This toxic combination of employment sorting and family clustering makes mobility more problematic than it needs to be. This Article closes by offering new ways of substantially altering employment sorting and family clustering to optimize the balance between the two and reap more benefits from mobility with fewer costs. These reforms would soften sorting while expanding clustering, and at the same time encourage certain forms of mobility (particularly to cities) that would permit a more optimal combination of sorting and clustering