Content area
Abstract
In 1970 the Canadian Journal of Economics published a paper by John Vanderkamp entitled, "The Effect of Outmigration on Regional Unemployment". Vanderkamp's results for the period 1951-61 indicated that for every five unemployed people leaving the Maritimes another two people became unemployed. The purpose of this thesis is to determine whether or not Vanderkamp's results are repeatable for the period 1971-81.
Vanderkamp's results are not repeatable for the period 1971-81 using Vanderkamp's model. Vanderkamp's model produces a negative employment multiplier with 1951-61 data and a positive employment multiplier with 1971-81 data. Possibilities examined for this result were that (a) the Maritime region was anomalous for 1971-81; (b) a few anomalous observations were distorting the results; (c) a feedback effect between outmigration and employment was distorting the results; and/or, (d) census divisions with net outmigration share characteristics that affect the relationship between outmigration and employment and that those characteristics are not shared by the census divisions with net inmigration.
Tests for (a), (b), and (c) proved negative. The test for (d) proved positive. For the period 1971-81, if Vanderkamp's model is expanded to include both census divisions showing net inmigration and net outmigration, his negative multiplier can be reproduced. (Abstract shortened by UMI.)
