NCJ Number
138969
Journal
Scottish Journal of Political Economy Volume: 39 Issue: 2 Dated: (May 1992) Pages: 213-228
Date Published
1992
Length
16 pages
Annotation
Empirical evidence is presented that supports a link between crime and unemployment in Scotland.
Abstract
Econometric problems posed by auto-correlation, heteroscedasticity and cross-sectional correlation are addressed within the feasible generalized least squares (FGLS) framework. Estimates based on a fixed effects model and an FGLS random effects model are also presented. Cross- sectional correlation in errors of crime rate equations is demonstrated in the study's regional-level data. In addition, the data do not discount the presence of both random and fixed effects across regions. The strongest finding is the extremely robust positive coefficient on the unemployment rate. Thus, the general finding is for a strong, well-determined positive relation between crime and unemployment rates. This relation, however, is not sensitive to the inclusion or exclusion of a public housing variable. Since unemployment cannot be dismissed as a determinant of Scotland's crime rate, policies designed to reduce regional unemployment in Scotland may also lead to a reduction in the regional crime rate. 13 references, 2 tables, and 1 figure