ABSTRACT
Migration and distance has been of scholarly interest ever since Ravenstein (1885) noted that most migrants proceed only a short distance. He thereby confirmed the opinion of A. Smith (1776 [1981]) that people constitute the most difficult form of capital to be transported over distance. Tellingly, Ravenstein also pointed out one of the key problems of migration research, in that migrants are defined as migrants when they cross administrative borders, but due to the varying size, shape, and situation of the areas, any analysis is fraught with difficulties:
[A] circumstance likely to lead to misconception, if not error, arises from the very unequal size of the counties. Rutland and Yorkshire are hardly comparable. A journey of 25miles at themost converts any native of Rutland into a “migrant,” whilst a native of Yorkshire to place himself into the same position might have to travel as many as 95 miles. (Ravenstein 1885, 168)
Since then, a number of studies have shown that there is a strong deterrent of distance to migration flows (see, e.g., Makower, Marschak, and Robinson 1938, 1939; Ha¨gerstrand 1957; Olsson 1965; Long, Tucker, and Urton 1988a, 1988b), and because migration is still almost exclusively defined by administrative borders, the problems identified over a century ago are equally valid today. Nonetheless, migration researchers have had to try to cope with these problems, and distance has come to occupy a central place in migration research.
