CRIMINOMETRIC ANALYSIS OF DETERRENCE HYPOTHESIS

This paper empirically estimates disaggregated crime categories for Sabah from 1968 to 2006. The criminometric analysis incorporated in a within sample analysis of cointegration and error correction framework and the beyond sample analysis using the decompositions of variance. Our findings suggest that any support for the deterrence hypothesis is sensitive to the inclusion of prison or courts related variables. In the long run we find that only robbery is exogenous in all crime model tested however, the beyond sample estimation proves that in longer time period of approximately 50 years the post-sample dynamic VDCs imply that a substantial portion of the variance of the forecast error of these crime are explained by their explanatory variables. The presence of cointegration between these variables tends to suggest that these economic and deterrent variables, along with alternative types of crime, are bound together by common trends. In other words, it can be said that although these cointegrated variables will have short-run or temporal deviations from their long-run common trends, eventually forces will be set in motion that will bring all the variables into states of equilibrium.