Abstract
It is argued that the view put forward by some analysts that unemployment had little or no effect on constituency results in the 1987 British general election is faulty. It rests on the failure to develop a satisfactory specification of the effects of unemployment, a problem which was also inherent in studies of previous British elections. A suggested new specification is tested and it indicates that changes in unemployment between 1983 and 1987 both at constituency and at regional level had a quite substantial effect on voting shifts in constituencies in 1987.
Get full access to this article
View all access options for this article.
